Upload
buique
View
219
Download
4
Embed Size (px)
Citation preview
Functional Modules in the flocculus of
the rabbit cerebellum
Fum::tional Modules in the flocculus of
the rabbit cerebellum
Functionele modules in de flocculus !Hill
het cerebellum LH:In het lconijn
PROHSCHR I FT
Ter verkrijging van de graad van doctor
aan de Erasmus Universiteit Rotterdam
op gezag van de rector magnificus
Prof. dr. P.W.C. Akkermans M. Lit.
en volgens besluit van het College van Dekanen.
De openbare verdediging zal plaatsVinden op
woensdag 15 december 1993 om 15.45 uur.
door
Tji11uw llok Marinus Tan
gelloren te Jogjalcortfl, lndonesie
I' ro motiecommiss ie
Promotoren: Prof. dr. J. Voogd Prof. dr. J.l. Simpson
Overige !eden: Prof. dr. H.J. Groenewegen
Dr. J. van der Steen
lndeH
General Introduction Zonal patterns in the cerebellum
Aim of the present study l. Chapter 1
"Anatomical compartments in the white matter of
the rabbit flocculus"
1.1 Introduction 1 . 2 Materials and methods
1.3 Results
page
1
1
5
7
7
1.3.1 The morphology of the flocculus and the paraflocculus 11
1.3.2 AChE-staining of the cerebellar cortex 15
1.3.3 Topography of the flocculus and floccular peduncle 16
1.3.4 Compartments and raphes in the white matter of 16
flocculus and paraflocculus
1.3.5 The floccular peduncle, topographical relations With
group and the cerebellar nuclei 2 7
1.4 Discussion 32 1.4.1 AChE and its distribution 32
1.4.2 Distribution of AChE in the cortex of the flocculus 34
1.4.3 Compartmental organization of the floccular white matter 38
1.4.4 Zones and compartments in the flocculus
1.4.5 Group y and the parvicellular part of
the lateral cerebellar nucleus
2. Chapter 2
"Zonal organization of the climbing fiber projection
to the flocculus and nodulus of the rabbit"
2.1 Introduction
2. 2 Materials and methods
2.3 Results
39
41
45
45
52
2.3.1 Projections from the caudal dorsal cap 52
2.3.2 Projections from the rostral dorsal cap and ventrolateral
outgrowth 65
2.3.3 The projection of the rostral pole of the medial accessory
olive and the dorsomedial cell column
2.4 Conclusions
2, 5 Discussion
2.5.1 The correlation of white matter compartments with the
climbing fiber zones in the flocculus
2.5.2 The correlation of white matter compartments with the
climbing fiber zones in the nodulus
2.5.3 The lateral A-zone and the nucleo-olivary pathway
in the rabbit
2.5.4 Functional considerations
3. Chapter 3
"Zonal organization of the floccule-vestibular projection
in the rabbit ..
3.1 Introduction
3.2 Materials and methods
3.2. 1 Retrograde studies
3. 2. 2 Anterograde studies
3.3 Results
3.3. 1 Retrograde experiments
3.3.2 Anterograde experiments
3.3.3 Conclusions
3. 4 Discussion
3.4. 1 The retrograde and anterograde tracing methods used
75
79
80
80
86
88
89
93
93
96
96
97
98
98
103
113
117
in this study 11 7
3.4.2 Zonal organization of Purklnje cells projecting to the
medial and superior vestibular nucleus and group y 119
3.4.3 Correspondence of cortico-vestibular w:l.th climbing
fiber zones in the rabbit flocculus
3.4.4 Functional considerations
4. Chapter 4
"Functional and anatomical organization of three-
125
126
dimensional eye movements in rabbit cerebellar flocculus 129
4.1 Introduction 129
4.2 Materials and methods 131
4.2.1 Eye movement recording 131
4.2.2 The coordinate system for measuring the eye movements 132
4.2.3 Electrophysiology 133
4.2.4 Histology 134
4.3 Results 134
4.3.1 Classification of eye movement responses 134
4.3.2 135° axis eye movement responses 134
4.3.3 Vertical axis eye movement responses 135
4.3.4 Other response types 137
4.3.5 Latencies of the response components 138
4.3.6 Anatomically delineated compartments 141
4.3. 7 Classes of eye movements in relation to
anatomical compartments
4.4 Discussion
4.4.1 Classes of evoked eye movements
4.4.2 Latency considerations
4.4.3 Topographical organization of the flocculus
4.4 .4 Comparison of eye movement and climbing fiber
143
149
!50
!53
!54
response axes 15 5
4.4.5 Comparison With eye movements evoked by floccular
stimulation in cat
5. General conclusions
6. AbbreViations
7. References
8. Summary
9. Samenvatting
10. Dankwoord
11. Curriculum Vitae
12. List of publications
!58
161
163
165
183
189
197
199
200
II an
pap en mam
General Introduction
Zonal patterns in the cerebellum
Many studies, using different anatomical and physiological
techniques, have aimed at subdiViding the cerebellum into distinctive
parts that exert control over specific motor systems for reflex and
voluntary movements, posture, and muscle tone. In view of the
homogeneity of the cerebellar cortex, delineation can not be based on
cytoarchitectonic differences as in the cerebral cortex. In the first half of
this century the macroscopic subdivision of the cerebellum into lobes,
lobules, and folia was used to distinguish functional units. The
comparative anatomical studies of Balk (1906) stressed the importance of
the subdivision of the mammalian cerebellum into vermis and
hemispheres. Bolk considered the vermis and hemispheres as
continuous, relatively independent "folial chains". separated by the
paramedian sulcus. Each of the folia! chains can be subdivided into
lobules, separated by deep, transverse fissures, and/or changes in the
direction of the folial chain. In the anterior lobe and the adjoining part of
the posterior lobe (the simple lobule) the transverse fissures continue
uninterrupted from the vermis into the hemisphere. The separation into
vermis and hemisphere is more complete in the caudal posterior lobe. In
some regions the cortex between vermis and hemisphere is completely
interrupted and white matter comes to the surface in the paramedian
sulcus. Bolk {1906) used the variation among species in the size of
different segments of the folial chains of vermis and hemispheres, in
combination with differences in motor behaviour, to relate parts of the
cerebellum to the control of specific motor behaviours. The comparative
studies by Larseil (1934, 1937) showed the constancy in the development
of the transverse lobular pattern throughout many mammalian species.
This approach eventually culminated in the subdivision of the cerebellum
into ten, transverse lobules, each of which was given a Roman numeral
(Larsen, 1952).
Other concepts on the functional organization of the cerebellum
I
were based on the connectiVity of different parts of the cerebellum. In the
early fourties a series of tracing studies using the Marchi method by
Jansen and Broda! (1940, 1942) demonstrated that each half of the
cerebellum could be subdiVided into 3 longitudinal zones that extended
perpendicularly to the transverse fissures and continued from one lobule
to the next. This subdiVision was based on the projections of the Purkinje
cells of the cerebellar cortex to their target nuclei. They distinguished a
medial (vermal), an intermediate and a lateral (hemispheran zone
projecting to the fastlgial, interposed, and lateral (dentate) cerebellar
nuclei, respectively. This longitudinal zonation concept was supported by
the findings of Chambers and Sprague (1955a, 1955b), who found that
lesions in different longitudinal zones resulted in different patterns of
deficiencies in motor behaviour, and they consequently concluded that
this zonation reflected a differentiated arrangement of motor control.
The longitudinal subdiVision was later found to be far more detailed.
In the sixties the collective results of a series of anatomical tracing and
myelin stain studies led Voogd (1964, 1969) to the conclusion that the
cerebellum was subdivided into 7 longitudinal compartments and zones.
By using the Hii.ggqvist stain (Hii.ggqvist, 1936), which reveals the axonal
myelin sheaths, Voogd (1964) demonstrated that the cerebellar white
matter is not homogeneous. In transverse sections, accumulations of thin
fibers (the so-called "raphes") were observed. They run as longitudinal
sheets across the transverse fissures and alternate with bundles of rather
coarse fibers. These raphes subdivide the cerebellar white matter into
compartments. each of which consists of a bundle of coarse fibers
delimited on either side by a narrow sheet of thin fibers (the raphes).
Recently, it has been shown that these raphes stain more densely for
acetylcholinesterase (AChE) than the compartments they delimit (Hess
and Voogd, 1986; Voogd et a!., 1987). An explanation for the preference
of AChE for the raphes has not yet been found.
The zonal organization of the cerebellar cortex is also reflected by
the presence in the molecular and Purkinje cell layer of a longitudinal
zonal arrangement of certain enzymes such as 5'-nucleotidase (Scott,
1963, 1964), and acetylcholinesterase (Maran! and Voogd, 1977).
2
neurotransmitters such as motilin (Nilaver et al., 1982) and taurin
(Chan-Palay et a!., 1982) and Purkinje cell markers such as zebrin mabQ
113 (Hawkes and Leclerc, 1987).
The compartments contain the olivocerebellar climbing fiber
afferents and the efferent axons of Purkinje cells that are distributed in
discrete. longitudinal zones, which extend perpendicular to the
transverse fissures. The axons from the Purkinje cells of a zone (or a
combination of zones) terminate in a specific cerebellar or vestibular
target nucleus. These Purkinje cells and their target nucleus both receive
afferents from a particular subnucleus of the inferior olive (Voogd, 1964.
1969: Groenewegen and Voogd, 1977: Groenewegen et al., 1979: Voogd
and Bigare, 1980). A reciprocal nucleo-olivary pathway connects the
target nucleus within this olivary subnucleus (Ruigrok and Voogd ( 1990).
The ensemble of a Purkinje cell zone (or combination of zones ). together
With its target nucleus, its climbing fiber afferents, and it nucleo-olivary
pathway constitutes an efferent cerebellar module (Voogd and Bigare.
1980).
The modular organization of the cerebellum is characteristic for its
output systems. but does not apply to the organization of its mossy fiber
input. Terminal fields of mossy fiber systems and the axons of their target
cells (the granule cells and the parallel fibers) are transversely oriented.
The mossy fiber input of the cerebellum, therefore, is closely related to
the transverse lobular subdtvision of the cerebellum advocated by Larsell.
The modular organization of the output systems of the cerebellar cortex
can be considered as an elaboration of the concepts of Balk and Jansen
and Brodal on the longitudinal subdiVision of the cerebellum.
StUdies of the function of longitudinal zones or cerebellar modules
are rare. The demonstration that acetylcholinesterase {AChE) stains the
borders of compiutments in the white matter of the flocculus (Hess and
Voogd, 1986) and the interest of Dr. J.l. Simpson, who occupied the
Tinbergen chair at the Erasmus University in 1989, in oculomotor control
by the flocculus prompted us to undertake a combined anatomical and
physiological study of compartments and Purkinje cell zones in the
flocculus of the rabbit.
3
According to Balk {1906; see also Voogd. 1975) the flocculus and
the paraflocculus are the last two segments of the folia! chain of the
hemisphere. Between lobules IX and X of the caudal vermis and the
paraflocculus and flocculus the cortex is interrupted and white matter
comes to the surface at the bottom of the paramedian sulcus and the
interparafloccular sulcus in the center of the folial loop of the
paraflocculus. The parafiocculus and flocculus constitute a curved, foliated
band that is delimited on its inner side by an extension of the white
matter in the paramedian sulcus and on its outer border by the white
matter of the cerebellar peduncles. The paraflocculus can be arbitrarily
subdivided into dorsal and ventral limbs, which are known as the dorsal
and ventral paraflocculus (Stroud, 1895). In some species a part of the
paraflocculus, known as the petrosal lobule, is lodged in the subarcuate
fossa of the petrosal bone.
The flocculus (Balk's uncus terminalis) is a well-demarcated lobule
that is located at the end of the folia! chain of the hemisphere. Its border
with the paraflocculus is formed by the posterolateral fissure. the earliest
fissure to appear during ontogeny (Larsell, 1937). In the depth of the
posterolateral fissure. the cortex of the flocculus and the paraflocculus are
continuous. The gross anatomy of the flocculus and paraflocculus of the
rabbit was described by Broda! (1940) and their morphogenesis was
traced by Larsen (1972). Yamamoto and Shimoyama (1977) gave a
detailed account of the foliation of the rabbit flocculus (fig. 0.1).
A zonal pattern has been identified for the climbing fiber afferent
input and efferent output of the flocculus by numerous anatomical studies
(see introduction of chapters 2 and 3} and physiological studies (see
introduction of chapter 4). However, the number and extent of the zones.
described in these studies vary considerably. Histological results from
different expertrrients were compared by the use of positioning relative to
the floccular folia. However, these folia show individual var1ation in size
and number, which complicates matching histological results from
different experiments.
4
Aim of the present study
To accurately compare histological results obtained in different
experimental paradigms a framework, independent of any individual
variation of the gross anatomy of the flocculus, is necessary. In chapter 1
AChE histochemistry is used to determine whether this technique shows
a consistent compartmentalization pattern in the white matter that could
then serve as the required independent framework. The distribution of
AChE in the floccular cortex and white matter, in the lateral cerebellar
nucleus and group y, and in the course of the floccular peduncle will be
described.
The possibility that the compartments in the white matter of the
flocculus correspond to anatomically and physiologically definable
functional modules was investigated in studies reported in chapters 2, 3,
and 4. The AChE chemoarchitecture of the floccular white matter, as
described in chapter l. could serve as an intrinsic framework that can be
used to correlate the position of labeled Purkinje cells and climbing fibers
with recording and stimulation sites in different sets of experiments,
This approach would allow the correct matching of histological data
obtained in anterograde (chapter 2) and retrograde tracing (chapters 3)
and microstimulation (chapter 4) experiments in the rabbit flocculus. In
chapter 2. studies on the origin, course and target zones of the afferent
climbing fiber input to the flocculus and nodulus are reported. Chapter 3
concentrates on the differential efferent projection of the floccular zones
to the vestibular and cerebellar nuclei. Chapter 4 considers the functional
significance of this differential efferent projection by describing the types
of eye movement elicited by stimulation of circumscribed compartments
in the floccular white matter.
5
fp
fa
fig. 0.1 Schematic drawing of a lateral view of the flocculus showing the most common configuration of folia labeled according to Yamamoto and Sh!moyama (1977).
6
Chapter 1
Rnolomical comportments in the white molter or the rabbit
flocculus.
1.1 Introduction
The subdivision of the cerebellum into longitudinal zones which
extend perpendicularly to the transverse fissures and continue from one
lobule to the next supplements the traditional subdivision of the
cerebellum into lobes. lobules, and folia. These zones have been defined
as longitudinal strips of Purkinje cells projecting to a particular cerebellar
or vestibular target nucleus {Voogd. 1964. 1969: Voogd and Bigare. 1980)
or by tbe origin of their olivocerebellar climbing fiber afferents (Courville
et al.. 1974: Groenewegen and Voogd. !977: Groenewegen et al.. !979:
Broda! and Kawamura, !980: Voogd, 1982: Bernard, 1987). A longitudinal
zonal arrangement also has been demonstrated for Purkinje cells
containing specific markers (Nilaver et al., 1982: Chan-Palay et al., 1981,
1982: Ingram et aL, 1985: Hawkes and Leclerc, 1987) and for the
distribution of enzymes, including acetylcholinesterase (AChE) in the
molecular layer of tbe cerebellar cortex (Scott, !963, 1964: Maran! and
Voogd, 1977a,b).
The zonal pattern of the cerebellar cortex is expressed in the
cerebellar white matter as a system of compartments that contain the
efferent axons and the climbing fiber afferents of the Purkinje cells of the
longitudinal zones {Voogd, 1969: Voogd, 1989). These compartments can
be distinguished with the Haggqv!st myelin stain as bundles of ratber
coarse Purkinje .:ell fibers separated by narrow strips of thin fibers, the
so-called "raphes" {Voogd. !964, 1969). Since then, it has been shown
tbat tbese raphes stain strongly for AChE (Hess and Voogd, 1986). The
co-localization of axons of the Purkinje cells of a zone and their climbing
fiber afferents in a particular white matter compartment and the
termination of these Purktnje cell axons and collaterals of their climbing
7
fiber afferents in one of the cerebellar or vestibular nuclei formed the
basis for the concept of the modular organization of the cerebellum
(Voogd, 1969: Groenewegen and Voogd, 1977: Groenewegen et al., 1979:
Voogd and Bigare, 1980). In this concept a module consists of a discrete
strip of Purkinje cells, its target nucleus and the olivocerebellar afferents
to these two structures. Anatomical and electrophysiological studies that
correlated climbing fiber afferents with the output of the cerebellar
cortex of the anterior lobe in the cat generally supported the modular
organization of this part of the cerebellum (Voogd, 1969, 1989:
Oscarsson, 1973: Andersson and Oscarsson, 1978a,b; Trott and
Armstrong, 1987a,b). Three compartments have been distinguished in
the white matter of the paraflocculus in Haggqvist-stained sections of the
cerebellum of cat and ferret (Voogd, 1964, 1969). The C2 compartment
is related to the posterior interposed nucleus, the D 1 and D2
compartments access different parts of the dentate nucleus. Climbing
fibers from the rostral medial accessory olive (MAO) use the C2
compartment to terminate on Purkinje cells of the C2 zone: the principal
olive projects to the D1 and D2 zones of the paraflocculus {Groenewegen
et al., 1979). Compartments in the white matter of the flocculus have
until now only been observed in the cerebellum of the monkey by Hess
and Voogd (1986) using AChE histochemistry.
Anatomical tracing and electrophysiological studies have shown that
the afferent climbing fiber and the efferent connections of the flocculus
are characterized by a zonal organization. Climbing fibers from the dorsal
cap (de), adjacent ventrolateral outgrowth (vlo) and the rostral medial
accessory olive (MAO) terminate in the flocculus in a pattern of
alternating longitudinal zones (Groenewegen and Voogd, 1977;
Yamamoto, 1979a: Gerrits and Voogd, 1982, 1989: Sato et a!., 1983:
Ruigrok et al. 1992). The climbing fiber zones distinguished by Yamamoto
(1979a) in the rabbit and by Sato et al. (1983) in the cat seem to coincide
with Purkinje cell zones projecting to specific vestibular and cerebellar
nuclei as delineated by the same authors (Yamamoto and Shimoyama,
1977: Yamamoto, 1978; Sato et al., 1982a,b). Seven and five climbing
8
fiber zones were distinguished in the flocculus of cat (Gerrits and Voogd,
1982) and rat (Ruigrok et a!. 1992). respectively.
According to Bolk (1906, see also Voogd. 1975). the flocculus and
the paraflocculus are the last two segments of the folial chain of the
hemisphere. Between lobules IX and X of the caudal vermis and the
paraflocculus and flocculus the cortex is interrupted and white matter
comes to the surface at the bottom of the paramedian sulcus and the
interparafloccular sulcus in the center of the folial loop of the
paraflocculus. The paraflocculus and flocculus thus constitute a curved.
foliated band of cortex, which is delimited on its inner side by an
extension of the white matter in the paramedian sulcus and on its outer
border by the white matter of the cerebellar peduncles. The paraflocculus
can be arbitrarily subdiVided into dorsal and ventral limbs, which are
known as the dorsal and the ventral paraflocculus (Stroud. 1895). In
some species a part of Lhe paraflocculus, known as the petrosal lobule, is
lodged in the subarcuate fossa of the petrosal bone.
The flocculus (Balk's uncus terminalis) forms a well-demarcated
lobule located at the end of the folial chain of the hemisphere. Its border
with the paraflocculus is formed by the posterolateral fissure, one of the
earliest fissures to appear during ontogeny (Larsell, 193 7). The gross
anatomy of the paraflocculus and the flocculus of the rabbit was described
by Broda! (1940) and their morphogenesis was traced by Larsell (1972).
The most detailed account of the foliation of the rabbit flocculus stems
from Yamamoto and Shimoyama (1977).
Some of the Purkinje cell zones projecting to the vestibular and
cerebellar nuclei (Yamamoto and Shimoyama. 1977: Yamamoto, 1978;
Bigare, 1980; Voogd and Bigare, 1980), and some of the climbing fiber
zones (Yamamoto, l979a), including the Cz zone which receives climbing
fibers from the rostral half of the medial accessory olive (Gerrits and
Voogd, 1982, 1989), extend across the posterolateral fissure into part of
the ventral paraflocculus. This part of the ventral paraflocculus is known
as the medial extension of the ventral paraflocculus (ME, Gerrits and
Voogd. 1982) in the cat and corresponds to folium p in the rabbit
(Yamamoto and Shimoyama, 1977).
9
For this paper the white matter compartmentation of the rabbit
flocculus was studied using Haggqvist's myelin stain and enzyme
histochemistry for AChE. Visualizing the borders of compartments with
AChE enzyme histochemistry is particularly easy and can be used to
compare the zonal distribution of the Purkinje cells with their climbing
and mossy afferents and to correlate electrophysiological and anatomical
data (Robertson and Logan, 1986; Tan et aL, 1989; Van der Steen et aL,
1991; see also Chapter 4).
1.2 Material and methods
Forty-five sectioned and AChE-stained brains of pigmented Dutch
belted rabbits were prepared. Some of these rabbits were also used for
axonal tracing or electrophysiological experiments to be reported in
subsequent papers. For AChE histochemistry the rabbits were deeply
anaesthetized with Nembutal and perfused intracardially first with l L of
saline at room temperature, followed by 2 L of a solution containing l-3o/o
paraformaldehyde and 0.5-1.25% glutaraldehyde in 0.1 M phosphate
buffer (pH 7.2-7.4) at room temperature, and finally with !0% sucrose in
0.1 M phosphate buffer (pH 7.2-7.4) at 4'C. The brains tern and
cerebellum were removed, rinsed in 10% sucrose phosphate buffer, and
embedded in a solution of 10% gelatin and 10% sucrose, which was
hardened in 10% formalin. The embedded tissue was stored overnight in
30% sucrose phosphate buffer, after which it was cut into 30 or 40 lilll
transverse, horizontal or sagittal sections and processed for AChE. We
used a standard thiocholine method to demonstrate AChE actiVity in the
cerebellum (Geneser-Jensen and Blackstad, 1971). Acetylthiocholine
acted as a substrate and ethopropazine as a blocking agent for
non-specific cholinesterases (Hess and Voogd, !986). Incubation Urnes of
6-9 hours were used to obtain optimal staining of the AChE-positive fibers
of the raphes. The sections were developed in a solution of 1 Oo/o
ferricyanide for 10 minutes, then rinsed in water, coverslipped and
examined With a light microscope.
Alternate sections for the Hii.ggqvist (1936) staining procedure
10
were stored in 10% formaldehyde for several weeks. After mounting, the
slides were immersed in a solution of 10% potassium bichromate and lo/o
calcium chloride at room temperature for 4-5 days, then rinsed in
distilled water and left in 96% alcohol for one day. Before staining in the
Haggqvist solution (methylblue and eosin, 1-24 hr) the slides were
immersed for 30 minutes in 10% phosphomolybdenic acid. A complete
set of transversely sectioned Haggqvist-stained, celloidin embedded 40
1ffi1 sections through the rabbit cerebellum was also used (see Voogd and
Fe!rabend (1981) for details on the Haggqvist stain). Three-dimensional
reconstructions of the flocculus and its white matter compartments were
prepared from styrofoam sheets or With the aid of the MacReco® program
(by E. Luitjens, University of Groningen) running on an Apple Macintosh
II computer.
1.3 Results
1.3.1 The morphology of the flocculus and the parajloccutus.
The flocculus is applied to the lateral surface of the middle
cerebellar peduncle. It consists of three to four rostrolaterally facing folia,
which are numbered fl-4 from ventral to dorsal (fig. 0.1). A short stretch
of cortex of fl that is folded back on the surface of the middle cerebellar
peduncle (folium m of Yamamoto and Shimoyama. 1977) serves as the
attachment of the roof of the lateral recess. Folia f3 and f4 are often fused,
as was the case in the cerebellum depicted in fig. l.l. Caudally the folia of
the flocculus taper Into a single folium. Caudally f3 and f4 taper and
disappear and folium m, fl. and f2 fuse Into a single folium. The
posterolateral ~issure demarcates the flocculus from the ventral
paraflocculus. This fissure runs obliquely along the medial side of .folium
f4 to the caudomedial pole of the flocculus, where it opens on the ventral
surface of the cerebellum.
ll
PF
©
PFLv with transected connection with PFLd
@
: between FLO and fp
floor offipl and tronsected connection with FLO
fi pi: inner surface
or layer
fissure be~ween f fmondf1
CD
!2
",-part of folium PFL1 3-4 removed to show
ventral pole of fp
The folium of the ventral paraflocculus bordering on the flocculus {folium
p of Yamamoto and Shimoyama. 1977) occupies a similar, oblique
position. The cortices of f4 (or the fused folia f3 and f4) and folium p are
continuous across the posterolateral fissure (figs. 1.2 nrs. 9-18). The
posterolateral fissure is rather wide in this region and a narrow strip of
cortex is found in the bottom between f4 and folium p (figs. 1.1 C; !.9 A.
C). More caudally, where the ventral paraflocculus starts to deviate
laterally, white matter coming to the surface in the posterolateral fissure,
separates folium p from the caudal pole of the flocculus (figs. !.1 D: !.2
nrs. 19-31). The cortex of the ventral paraflocculus covers the ventral
surface of the parafloccular stalk and forms a rosette of three folia on the
ventrocaudal aspect of the petrosal lobule, which in the rabbit consists of
the entire parafloccuius. The two rosettes on the dorsal and caudomedial
aspect of the petrosal lobule belong to the dorsal paraflocculus (fig. 1.1 A).
The folial chain of the paraflocculus and flocculus is reflected on the
ventrolateral surface of the cerebellum. As a consequence, the media
lateral position of these lobules is reversed when compared to the
proximal part of the chain. With their outer border facing medially.
Caudally, the cortex of the paraflocculus is continuous with the cortex of
the paramedian lobule. More rostrally, white matter extending from the
surface of the middle cerebellar peduncle separates the cortex of the
paraflocculus and the flocculus from the anterior lobe and the simple.
ansiform and paramedian lobules (asterisk in fig. 1.1 B).The white matter
that borders the cortex of flocculus and paraflocculus on the lateral side is
an extension of the white matter at the bottom of the paramedian sulcus
fig. 1.1 Drawing of a computer-reconstruction of the rabbit's right flocculus and paraflocculus. This specimen is also depicted in fig. !.9. Only the molecular layer was reconstructed. A) Rostrodorsolateral view. Flocculus and ventral and dorsal paraflocculus are indicated with different symbols. For border between dorsal and ventral paraflocculus see text and fig. !.6. BJ Photograph of the rabbit cerebellum. Rostrodorsolateral view. Asterisk indicates white matter on dorsal surface of parafloccular stalk. CJ The ventral paraflocculus, same view as in A. D) Dorsal view of the flocculus and the ventral paraflocculus. E) Ventral View of the flocculus and the ventral paraflocculus.
13
14
betv.reen vermis and hemisphere. It extends from lobules IX and X of the
caudal vermis, along the caudal border of the paramedian lobule over the
petrosal lobule, where it radiates over the caudomedial surface of the folia
of the flocculus and paraflocculus.
1.3.2 AChE-staining of the cerebeUar cortex.
The granular layer of the cerebellar cortex of the rabbit stains more
intensely for AChE than the molecular layer. The two layers are separated
by the AChE-negative Purkinje cells. The staining in the granular layer is
irregular with strongly reactive patches scattered between less reactive
granule cells and neuropil. The AChE-positive patches were identified as
glomeruli and Golgi cells, first by Gerebtzoff (1959) and later by many
others (Csillik et al., 1963; Altman and Das, 1970; Brown and Palay,
1972}. Generally the staining increases in the superficial part of the
granular layer. next to and in between the Purkinje cells. Strong and
more uniform AChE staining is present in the entire granular layer of the
flocculus. The staining of the granular layer of folium p and its transition
into the flocculus is particularly dense (fig. 1.3 D). Intense staining of the
granular layer is also present in the vermis, in the depth of the fissures
and in lobules IX and X. where it extends farther apically. The molecular
layer of the flocculus and the ventral paraflocculus can be subdivided into
hvo layers. The lower one third of the molecular layer, next to the
Purkinje cell layer, is more densely stained than the superficial two
thirds.
fig. 1 .2 Schematic drawing of 31 transverse sections through the flocculus and ventral paraflocculus showing the location and course of AChE-positive raphes. The white matter compartments (Cz, FC 1-FC4) and cerebellar nuclei are indicated with different symbols. Arrow points at the posterolateral fissure. Between the caudal pole of the flocculus and folium p the cortex is interrupted.
15
The AChE staining in the lower one third disappears abruptly at the
proposed border of the ventral with the dorsal paraflocculusl .(fig. 1.6
A, arrow). The molecular layer of the dorsal paraflocculus is poor in AChE.
Medium-sized AChE positive cells are found scattered over the lower half
of the molecular layer in all lobules of the rabbit cerebellum {fig. 1.6 B. C).
According to Spa9ek et al. {1973) these AChE-stained cells correspond to
the displaced Golgi cells of Cajal {1911).
1.3.3 Topography of the floccuLus and thefloccuLar peduncLe.
In sections incubated for AChE a narrow strip of intensely stained
fibers marks the border between the white matter of the flocculus and
the middle cerebellar peduncle [figs. 1.2: 1.3: l.lO: 1.11: 1.13).
Ventrally, the flocculus borders on the cochlear nuclei with the granular
layer of its most ventral folium (folium fl or m, figs. 1.2; l.3 nrs. 4-29
and 1.3). The border between the dorsal cochlear nucleus and the
granular layer, both of which react strongly for AChE, is sometimes
difficult to establish. More caudally, where the medial cerebellar peduncle
has entered the cerebellum, the fibers of the flocculus assemble in the
floccular peduncle located between the flocculus and the restiform body,
with the lateral cerebellar nucleus on its dorsal side and the dorsal
cochlear nucleus more ventrally {figs. 1.2 nrs. 25-31: 1.6). The cells and
the neuropil of group y, which stain strongly for AChE, appear in this
area. Still more caudally the entire complex of the floccular peduncle and
group y arches medially over the restiform body and merges with the
superior and medial vestibular nuclei (fig. l. 7 B-D). The floccular
peduncle and group y are located next to the cochlear nuclei and the
dorsal acoustic stria, which follow a similar curve over the restiform body
at a slightly more ventral and caudal level {fig. l. 7 C-E).
1 A dorsal and a ventral paraflocculus have been distinguished in !be cerebellum of the rabbit by several authors
(Brodal, 1940; Larsell, 1970), but the precise border between the lobules was never indicated. It is proposed that
this border is siruated at the transition in the chemoarchitecture of the molecular layer of the two division of t:lle
para-flocculus (see fig. 1.6).
16
1.3.4 Compartments and raphes in the white matter of the flocculus and
paraflocculus.
AChE histochemistry reveals narrow strips of densely stained fibers
in the white matter of the cerebellum (figs. 1.2; 1.10; 1.11; 1.13). The
strips separate more lightly stained fiber compartments. In serial
sections the AChE-positive strips can be traced as sheets throughout the
white matter of the flocculus and the paraflocculus (fig. 1.3). In a drawing
of a computer reconstruction of the white matter of the flocculus the
compartments appear as oblique slabs that taper towards the caudal pole
of the flocculus (fig. 1.5). AChE positive fibers of the raphes coincide with
accumulations of dark blue. small diameter fibers found in Haggqvist
stained sections. at the borders (the so-called raphes) of white matter
compartments containing larger diameter fibers (fig. 1.4). At the sites
where the raphes meet the granular layer, this layer often stains more
intensely. At these points the granular layer also shows a characteristic
bulging into the white matter. These bulges form strictures in the white
matter. In the vermis and hemispheres, concentrations of AChE-positive
glomeruli are obvious in the midline and at certain parasagittal positions.
Such a regular glomerular staining pattern is not seen in the flocculus or
the paraflocculus.
In the white matter of the flocculus four raphes usually can be
distinguished. They subdivide the white matter into five compartments
(figs. 1.3; 1.5). The most lateral compartment corresponds to the C2
compartment ofVoogd (1964. 1969). The other four are numbered from
lateral to medial and indicated as (floccular compartment) FC 1 -FC4 The
raphes are named after the compartments they border.
Raphe C2/l. the most lateral and one of the most intensely staining
AChE raphes, can be traced from the white matter of the paraflocculus
into the flocculus. It delimits a white matter compartment located on the
caudal side of the stalk of the paraflocculus. In horizontal and sagittal
sections (fig. 1.9; 1.12) reveal how this raphe becomes applied to the
caudal border of the lateral cerebellar nucleus and continues as the
border between the lateral and posterior interposed nuclei {fig. 1. 7
17
18
fig. I .3 Photomicrographs of four AChE-stained transverse sections (A-DJ through the left flocculus. corresponding respectively to sections 4, 11, 16. and 27 of figure 1.2. Small arrows indicate AChE-positive raphes between the compartments.
19
fig. 1.4 Photomicrographs of a Haggqvist-stained transverse section through the flocculus (A) and a high power view of the border area between compartments FC2 and FC3 (B). The level of the section approximately corresponds to fig. 1.3 C. Note the relatively coarse fibers in FC2. Bar in A = 500 ~m. and in B = 75 ~m.
E-F). The caudal compartment of the paraflocculus, which contains the
posterior interposed nucleus, therefore, can be identified as the Cz
compartment (Voogd, 1964, 1969). The C2 compartment continues from
folium p into the white matter of the caudal pole of the flocculus, from
where it can be .traced for a short distance into the white matter of fl-3
(figs. 1.2; 1.3; 1.7; 1.9-1.13). The large, rostral compartment of the
paraflocculus supposedly corresponds to the D-compartment, but it does
not show clear signs of a subdivision into D 1 and Dz compartments as
present in cat, ferret and monkey (Voogd, 1964, 1969, Voogd eta!, 987,
see, however. asterisk in fig. 1.11 A). Compartment C2 stains rather
20
fig. 1.5 Drawing of a rostrolateral view of a computer reconstruction of the flocculus (A), the cortex and the five white matter compartments (B) and the white matter compartments pulled slightly apart (C). Note the oblique position of the compartments. In the caudal flocculus compartment FC4 has disappeared. and FC 1 and FC3 cover the FCz compartment. The reconstruction was generated from the sections depicted in fig. 1.2. The numbers in B refer to these sections.
densely for AChE. Compartments FC!. FCz. and FC3 are present in all folia
of the flocculus and In folium p. FCz. FC3. and FC4 appear in the
rostralmost sections of the transverse series depicted in fig. 1.2, FC 1
appears next and C2 is only present in the caudal half of the flocculus.
When traced caudally the compartments shift medially and ventrally (figs.
1.2; 1.3; 1.5).
Raphe l/2 is distinct and delimits FC! on its lateral side. FCz is
21
fig. 1.6 Photomicrograph of an AChE-stained transverse section through the dorsal and ventral paraflocculus and high power views of the cortical layers of the dorsal (B) and the ventral paraflocculus (C). Note the abrupt change in staining density of the inner third of the molecular layer of the dorsal paraflocculus at its border with the ventral paraflocculus {large arrow in A) and the displaced Golgi cells in the molecular layer {triangles in B and C). Small white arrows indicate the raphe bordering the Cz compartment.
characterized b~ its pale appearance in comparison with the adjoining
compartments FCt and FC3 (fig. !.3 C; !.5). From Haggqvist·stained
sections the impression is gained that the fibers contained in the FC2
compartment are of a slightly larger calibre than those in the adjoining
compartments (figs. 1.3 C; 1.4 A). The number of large fibers decreases
and the amount of AChE increases in the lateral half of FCz, towards the
22
intensely stained raphe 1/2 (figs. 1.4 A; l.!O.A). Raphes l/2 and 2/3 fuse
in the caudal half of the flocculus, dorsal to FC2, which now has the shape
of a triangle, With its base resting on the granular layer of folium fl and its
apex pointing dorsally. The point of fusion of the two raphes around the
apex of FC2 can always be recognized by intense staining for AChE (figs.
1.4 A; 1.10 A). The fused raphe l/3 extends dorsally as the border
between FC1 and FC3 and terminates at the granular layer of folium p.
(fig. 1.2 nrs. 11-15; 1.3 B).
Compartment FC3 is located between the raphes 2/3 and 1/3
laterally and raphe 3/4 and the medial border of the flocculus With the
middle cerebellar peduncle medially. Raphe 2/3 stains less intensely for
AChE than the other raphes, but can be identified because it separates the
dense staining in FC3 from the pale area of FC2. Compartment FC4 is
narrow and located parallel to the border of the flocculus Vlith the middle
cerebellar peduncle. FC4 can be distinguished as a separate compartment
only in the rostral one-third of the flocculus (figs. 1.2 nrs. 1-11; 1.3 A;
1.10 C; 1.13 B). Raphe 3/4 fuses With the medial border of the flocculus
and the fibers of FC4 disappear into a meshwork of AChE-positive cells,
loca1:ed between the cochlear nuclei, the middle cerebellar peduncle, and
the restiform body (figs. 1.2 nrs. 8-11; 1.3 C. D).
Compartment FC1 straddles the posterolateral fissure over most of
its extent and continues for some distance into the white matter of folium
p (fig. 1.2 nrs. 9-23). FC2 is only represented In the rostral pole of folium
p (fig. 1.2 nrs. 9-ll). FC3 IS narrow in the rostral white matter of folium
p and Widens more caudally (fig. 1.2 nrs. 9-15). Its caudal extent and its
relation With the D compartment of the paraflocculus could not be
determined.
An additional raphe, which subdivides compartment FC1, is
sometimes present in an intermediate position between the intensely
stained raphe Cz/1 and the raphe 1/2, which itself can be recognized by
its fusion with raphe 2/3. This raphe is present in the horizontally
23
®
\
'· ,r __
® PMD
Pfld
@ 2mm P\ I"
Hg. 1. 7 Drawihg of AChE-stained transverse sections through the cerebellum showing the close relationship between flocculus, floccular peduncle, group y and the cerebellar and vestibular nuclei. Note the extent of compartment C2. For explanation of symbols used to indicate floccular compartments and cerebellar nuclei see fig. l.l2. The distribution of AChE in deep parts of the molecular layer is indicated by stippling.
24
fig. I .8 Photomicrographs of adjacent transverse AChE- (A) and Nisslstained (B) sections through group y and floccular peduncle. Small, densely ACHE-stained cells constitute the ridge indicated With r. Ventral group y (vY) consists of small cells (mean 24 x 6 ~ml in a densely AChE-stained matrix; the cells of dorsal group y (dY) are larger (mean 32 x 12 ~m) and located between the fibers of the floccular peduncle. Lpc contains small neurons (mean 12 x 12 )J.m). and the lateral cerebellar nucleus contains large cells (mean 26 x 22 )lm).
25
®
26
sectioned specimen of fig. 1.9 D.E (arrows) and 1.10 C (small white
arrows), but it cannot be recognized in the transverse sections of the case
illustrated in figs. 1.2 and 1.3.
1.3.5 The floccular pedunde, topographical reiations with group y
and the cerebeUar nuclei.
The fibers contained in the compartments of the flocculus enter the
cerebellar nuclei and the floccular peduncle, but in this process some of
the borders between the compartments can no longer be distinguished.
The unique relation of the Cz compartment with the posterior interposed
nucleus has been described above. Caudally, raphe l/3, dorsal to FC2.
becomes less distinct (fig. 1.2 nrs. 23-25). Fibers of FC1. FCz and FC3
enter the ventral region of the lateral cerebellar nucleus and/ or shift
medially to constitute the incipient floccular peduncle in the area
ventromedial to the lateral cerebellar nucleus, dorsal to the cochlear
nucleus and medial to the restiform body. The conglomerate of fibers
travelling in different directions imparts a patchy appearance to the
AChE-staining in caudal compartment FC3 (fig. 1.3 C). Much reduced
compartments FC 1 and FC2, flanked by the C2 compartment, can be
traced through the caudal pole of the flocculus, into the roof of the lateral
recess (figs. 1.3 D; 1.13 B), and farther along the attachment of the roof
of the fourth ventricle in the direction of lobule X of the caudal vermis. At
this point. the border of compartment FC2 with the floccular peduncle is
formed by an AChE-positive cellular ridge, which may be continuous with
the dorsal cochlear nucleus (r in figs. 1.3 D: 1.7 B,C: 1.11 A: 1.13 C).
Similar small, AChE-positive cells invade the raphes and compartments
in the caudal pole of the flocculus and the incipient floccular peduncle.
fig. 1.9 Drawings of AChE-stained horizontal sections through the flocculus. A is the most dorsal, G the most ventral section. Compartment FC 1 is subdivided by an additional raphe in medial and lateral parts (arrow in D and E). For explanation of symbols indicating compartments and cerebellar nuclei see fig. 1.12.
27
28
Group y can be considered as the bed nUcleus of the floccular
peduncle. Dorsal and ventral subdivisions of group y can be distinguished.
The dorsal division consists of both strands and single, AChE-positive
neurons, lying between the fibers of the floccular peduncle, ventral to the
panricellular extension of the lateral cerebellar nucleus (fig. 1.9). These
fusiform cells are much larger than those of ventral group y (see legend
fig. 1.8). The small cells of the ventral subdivision constitute a dense,
AChE-positive cap over the restiform body at its border "With the floccular
peduncle. The floccular peduncle and the cells of group y arch over the
restl.form body, ventral to the parvicellular extension of the lateral
cerebellar nucleus. to merge with the rostral pole of the superior
vestibular nucleus (fig. l.!O E).
The relations and the composition of the floccular peduncle are
most distinct in horizontal and sagittal sections. Laterally, the peduncle is
separated from the FC2 compartment in the caudal pole of the flocculus
by the AChE~positive cellular ridge (figs. 1.9; 1.12 D). This ridge still
separates the peduncle from a narrow offshoot from FC2, located in the
wall of the lateral recess and penetrating betvveen group y and the dorsal
cochlear nucleus (figs. !.12 E,F; !.13 C,D). Before the peduncle enters
the caudal pole of the superior vestibular nucleus, it divides into rostral
and caudal bundles. The caudal bundle stains more intensely for AChE and
corresponds to LOwy's (1928) bundle (!.9 C; !.13 E.F: 1.14 C,D). It
passes the lateral angle of the fourth ventricle, where it is located dorsal
to the medial vestibular nucleus and turns rostrally. The rostral bundle of
the floccular peduncle becomes lost in the superior vestibular nucleus.
fig. 1.1 0 Photomicrographs of three horizontal sections through the flocculus. In the dorsalmost section (A) compartment FC2 has disappeared from the middle part of the flocculus and the !/2 and 2/3 raphes have fused. Note the presence of an additional raphe (indicated with an arrow in C) which subdiVides FC 1 into lateral and medial halves. Compare fig. 1.9 D. E.
29
rostral t ~ lateral
30
The subdivision of the cerebellar nuclear complex in the rabbit can
best be appreciated from the illustrations of horizontal and sagittal
AChE-stained sections. The lateral cerebellar nucleus is fairly compact
with a lateral tail extending into the stalk of the paraflocculus and a
medially-directed hilus. This nucleus consists of two groups of large
neurons in an AChE-poor neuropil in the rostromedial and caudolateral
parts of the nucleus, separated by an area With somewhat smaller cells in
a strongly AChE-reactive neuropil (figs. !.3 D; !. 7; 1.9; 1.13). Like most
neurons of the cerebellar nuclei they contain AChE. The caudolateral
group of large cells constitutes the lateral tail and the caudal pole of the
lateral nucleus. Medially, the lateral nucleus borders on the anterior
interposed nucleus. Fibers from the white matter of the flocculus enter
the ventral aspect of the lateral nucleus and traverse its AChE rich border
region (figs. 1.2 nr. 28-31; 1.3 D; 1.7 B,C). These fibers separate the
ventromedial extension of the lateral nucleus, which contains small- and
medium-sized neurons in a moderately AChE-reactive neuropil, from the
lateral and caudal magnocelluiar shell of the lateral nucleus. This
subnucleus is indicated as the parvicellular part of the lateral nucleus
(Lpc, fig. 1.8 A). The floccular peduncle abuts the Lpc on its ventral and
rostral side. A conspicuous AChE-positive patch of small round cells is
located along the caudal and dorsal border of the Lpc (r in figs. 1.8: I. 9
B; 1.12 D,E; 1.13 C). This cell group may be continuous With the AChE
positive ridge separating the floccular peduncle from the flocculus.
Medially, the Lpc joins the vestibular nuclei, while ventrally it merges
With the fibers of the floccular peduncle and group y.
fig. 1.11 Photomicrographs of a horizontal (A) and a sagittal (B) section through the flocculus and paraflocculus. Note position of C2 compartment in the parafloccular stalk in A and B. A densely AChE-positive cellular ridge (r) separates the caudal white matter of the flocculus from the floccular peduncle (pf + y) in A. Asterisk in A indicates a vague raphe subdiViding compartment D.
31
The anterior interposed nucleus is located on the medial side of the
lateral nucleus. It can be distinguished from the latter by its
medium-sized neurons in a pale. non-AChE reactive neuropil. Lateral to
the superior cerebellar peduncle it extends far ventrally, bordering on the
highly AChE-reactive superior vestibular and parabrachial vestibular nuclei
(figs. 1. 7 B-E; L9 A-D). The posterior interposed nucleus is found in a
rather medial and caudal position. Its ventrolateral pole reacts strongly
for AChE and is sharply delimited by a fiber lamella from the lateral
nucleus (figs. 1. 7 F: 1.9 A: 1.12: 1.13). The dorsal part of the nucleus
contains large cells in a pale neuropil. A meshwork of cells separates the
posterior interposed nucleus from the medial cerebellar nucleus. This
meshwork is traversed by the efferent fibers of the posterior interposed
nucleus. which continue as the medial one third of the brachium
conjunctivum, and by bundles of corticovestibular {perforating) fibers.
1.4 Discussion
1.4.1 AChE and its distribution.
AChE is present in neuronal as well as in non-neuronal tissues
(Silver 1974). In the brain, AChE is associated with the hydrolysis of
acetylcholine in the synaptic cleft of axonal terminals of cholinergic
neurons and "cholinoceptive" cells. Other possible functions of AChE in
the brain have been discussed by Greenfield ( 1984) and Appleyard and
Jahnson {1992). Pseudocholinesterase, which converts other substrates
(Maran!, 1982), is present in endothelial and glial celts (Barth and
Ghandour. 1983}, but was inhibited in our experiments by ethopropazine.
fig, 1.12 Drawings of AChE-stained sagittal sections through the left flocculus, showing compartments in the flocculus, parafloccular stalk. and emerging floccular peduncle. Compare photomicrographs of figs. l.ll B and 1.13.
32
~ L I LAChE -rich • sv .
FC1 .
Lpc Ed FC 2 MV
~ IP I I PAChE-rich a FC3 ~ DV .
lA ~ FC, ~g~?g?i LV
E~IJ Fl FAChE-rich • c, - AChE AChE in the lower molecular layer
33
Recently, pseudocholinesterase was derrionstrated in certain
Purkinje cells in lobules IX and X of the cerebellum of the rat, but the
flocculus was not investigated (Gorenstein et al. 1987). In this paper,
AChE was distributed in sheets of thin fibers at the borders of
compartments in the white matter of the flocculus and the paraflocculus. AChE-staining Within the compartments shows characteristic differences.
Compartment FC2 lacks AChE, especially medially where large diameter
fibers accumulate. The other compartments, particularly C2. stain more
intensely. The subdiVision of the white matter into compartments is not
reflected in a zonal distribution of AChE over the cortex of the flocculus
and the parafiocculus.
1.4.2 Distribution of AChE in the cortex of the jlocculus.
Strong staining for AChE, with the highest reactivity in patches
between less reactive granule cells, is observed in the entire granular
layer of the flocculus and folium p. The deep stratum of the molecular
layer also reacts intensely for AChE, but this type of staining extends beyond the flocculus and folium p to include (by our definition) the entire
fig. 1.1::1 Photomicrographs of sagittal sections through the flocculus. A} Lateral section through folium p and the lateral pole of the flocculus with the compartments FC1 -FC3. Arrow indicates the raphe bordering the c2 compartment. B) More medial section. The incipient floccular peduncle is separated from compartments FC1, and FC2, and C2 in the white matter of the flocculus by a densely AChE-stained ridge of small cells (r). FC3 joins the floccular peduncle and FC4 appears in the rostral flocculus. C) Section through the lateral pole of the posterior interposed nucleus (IP). The floccular peduncle is divided into the AChE-rtch bundle of Uiwy (16) and a rostral, AChE-poor portion, which contains the neurons of dorsal group y. Note the densely stained ridge of small cells (r) dorsal to the parvicellular lateral nucleus. D) Higher magnification of the floccular peduncle and the ventral and dorsal group y.
34
dorsal
A
B
35
ventral parallocculus. Strong staining of the granular layer is also observed
in the cortex of the vermis at the bottom of the fissures. This staining
extends further apically in the ventral part of the anterior lobe and the
caudal vermis, including the entire lobule X and the ventral lobule IX. In
other parts of the cerebellum staining Is most Intense In the superficial
granular layer. next to the Purkinje cells, which always are left unstained.
AChE is also present in the deep stratum of the molecular layer of the
vermis, the hemisphere of the anterior lobe, and the simple and
paramedian lobules. More laterally, the molecular layer of the
hemisphere, including the ansiform lobule and the entire dorsal
paraflocculus, Is entirely AChE- negative.
According to many authors the patchy distribution of AChE in the
granular layer is due to staining of cerebellar glomeruli and Golgt cells
(Gerebtzoff 1959: Csillik et al .. 1963: Brown and Palay, 1972: Altman and
Das, 1970). Glomerular staining for AChE may be indicative of cholinergic
transmission because choline-acetyl transferase (ChAT, the synthesizing
enzyme of acetylcholine) immunoreactive mossy fiber rosettes were found in the granular layer In rat, guinea pig, cat, rabbit, monkey and man (Kan.
1978, 1980: Illing, 1990: Ojtma et al., 1990; Ikeda et al., 1991: Barmack
et al., 1992a: De Lacalle et al .. 1993). In the rat a plexus of thin, beaded
ChAT-immunoreactive fibers pervades all layers of the cortex. ChAT
immunoreactive mossy fibers were mostly restricted to lobules X and
ventral lobule IX and the flocculus (Ojima et al., 1990: Barmack et al ..
l992a). The plexus of beaded ChAT-immunoreactive fibers could not be
detected in the cerebellum of the rabbit. In this species, ChAT-containing
mossy fibers were present in lobules I and II of the anterior lobe and the
portions of lobules X and IX facing the posterolateral fissure. They were
rather scarce in the flocculus. but the ventral paraflocculus of the rabbit
was densely innervated {Barmack eta!., 1992a).
The resemblance between the distribution of the vestibulocerebellar
mossy fiber projections (Gerrits et al., 1989: Epema eta!., 1990: Tan and
Gerrits, 1992: Thunnissen eta!., 1989) and the regions with strong AChE
reactiVity of the granular layer in the rabbit is very close indeed. Barmack
et a!. (l992b) traced the cholinergic mossy fiber projections to lobules X
36
and IX and the flocculus in rat and rabbit from ChAT-immunoreactive
neurons in the caudal medial vestibular nucleus and the nucleus
preposttus hypoglossi with a double-labeling HRP retrograde axonal
transport-immunohistochemical technique. A projection to the ventral
paraflocculus of the rabbit took its origin from ChAT -immunoreactive cells in the nucleus prepositus hypoglossi.
AChE staining of the molecular layer is more variable among
different species and is therefore more difficult to interpret (Friede and
Fleming, 1964; Silver, 1967, 1974; Karczmar. 1969; Maran! and Voogd,
1977a.b: Maran!, 1982, 1986; Hess and Voogd, 1986). The presence of
AChE in the deep stratum of the molecular layer of the flocculus and
ventral paraflocculus and its absence (by definition) in the dorsal
paraflocculus of the rabbit does not appear to be related to the
distribution of cholinergic terminals, which could not be demonstrated in
the molecular layer of the rabbit cerebellum (Barmack et al., 1992a). The
AChE-positive displaced Golgi cells of Cajal (1911) and Spa9ek et al.
(1973) cannot be responsible for the differential distribution of AChE over
the molecular layer of the ventral and dorsal paraflocculus of the rabbit
cerebellum. because these cells are distributed ubiquitously over the
cortex. ChAT-immunoreactive Golgi cells, some of which of which were
found in the cat (Illing, 1990; Ikeda et a!., 1991), were not observed in
the rabbit (Kan et al., 1978; Barmack et al .. 1992a). The distribution of
nicotinic cholinergic receptors (Hunt and Schmidt, 1978) and
muscarinic (M2) receptors (Mash and Potter, 1986) over the granular
layer of the rat cerebellum corresponds well with the punctate AChE-staining and the ChAT-immunoreactivity of the mossy fiber
rosettes. Muscarinic receptor distribution over the molecular layer is
more variable, with multiple bands in the molecular layer of the caudal
vermis, flocculus·. and paraflocculus of the rabbit, less distinct zonation in rat and guinea pig and a uniform distribution in the mouse (Yamamura and
Snyder, 1974; Kuhar and Yamamura, 1975: Gulya and Kasa. 1983; Mallo]
et al., 1984: Rotter et al, 1979a,b; Neustadt et al., 1988).
37
1.4.3 Compartmental organization of the floccular white matter.
An AChE subdivision of the cerebellar white matter has previously
been observed in the anterior lobe of the cat (Voogd et al.. 1986) and in
the cerebellum of the monkey. including the flocculus and paraflocculus
(Hess and Voogd. 1986; Voogd et al.. 1986; Voogd et al.. 1987). Both in
the cerebellum of cat and monkey. and in the flocculus of the rabbit, the
AChE-positive fiber bundles correspond to accumulations of thin
myelinated fibers (raphes) in Haggqvist-stained sections. Raphes usually
separate bundles of coarser Purkinje cell axons arising from longitudinal
cortical zones on their way from the cortex to their target nuclei.. The
nature of the AChE-positive fibers of the raphes is puzzling. They may
represent AChE-positive mossy fibers, which have become concentrated
at the borders of sheets of Purkinje cell axons haVing a different
destination. A possible correlation between concentrations of AChE
positive mossy fiber rosettes in the granular layer and the presence of
fascicles of strongly stained fibers in the white matter was noticed by
Boegman et al. (1988) for the vermis and paravermis of the rat
cerebellum. These authors described a close correspondence between
the localization of bands of MAB-Q 113 (zebrin) immunoreactive Purkinje
cells and the zonal AChE pattern. MAB-Q 113 positive Purkinje cells in the
flocculus of the rabbit are not distributed in a clear zonal pattern, but
rather in patches; moreover MAB-Qll3 immunoreactiVity was absent
from the axons of Purkinje cells in the white matter of the flocculus (Tan
et al., 1989). Another factor that may determine the staining of the
raphes of the flocculus is their invasion by small, strongly AChE-positive
cells. Concentrations of such cells are found as a ridge at the lateral
border of the floccular peduncle. where it separates from the flocculus.
and as strongly AChE-reactive patches. dorsal to the parvicellular lateral
cerebellar nucleus. The scattered. AChE-positive cells in the white matter
of the flocculus may correspond to part of the basal interstitial nucleus of
primates (Langer. 1985). The AChE-positive neurons of the floccular
white matter and group y of the rat are ChAT-negative (Komei et al.,
1983).
38
Four compartments were tentatively identified in the white matter of the flocculus and the ventral paraflocculus of macaques (Hess and
Voogd 1986: Voogd et a! .. 1987). Both in the monkey and in the rabbit
the most lateral C2 compartment extends into the petrosal lobule and the
dorsal paraflocculus and includes the posterior interposed nucleus.
Rostrally, compartment C2 continues into the ventral paraflocculus in
both species. No equivalent of compartment FC4 was found in primates.
1.4.4 Zones and compartments in the flocculus.
Anatomical as well as physiological studies have Indicated the
presence of longitudinal zonation in the flocculus. In an anterograde
tracing study in the cat Gerrits and Voogd (1982) showed that climbing
fibers from the dorsal cap and the adjoining ventrolateral outgrowth, the
bend of the principal olive, and the rostral pole of medial accessory olive
terminate in longitudinal strips in the molecular layer. In the flocculus
and the adjacent folia of the ventral paraflocculus. they observed seven
climbing fiber zones (Fl-F6, C2l In the molecular layer derived from
different parts of the inferior olive. An additional zone (F7) was observed
in the adjacent ventral paraflocculus. The fibers travelled in bundles in
adjacent longitudinal white matter compartments. However, they did not
correlate the location d these fiber bundles to matching AChE or
Haggqvist-stained material. A lateral C2 zone, receiving climbing fibers
from the rostral pole of the medial accessory olive was clearly present in
the cat, and the fusion of the F2+3 and the F5+6+ 7 zones around F4 (a
zone which receives climbing fibers from the caudal part of the dorsal
cap, Gerrits and Voogd 1982) may correspond to the fusion of
compartments FC1 and FC3 In folium p of the flocculus of the rabbit. The
topographical relations of the five climbing fiber zones that were traced
by Ruigrok et a!. (1992) With anterograde transport of Phasaeolus vulgaris
lectin to the flocculus and the paraflocculus of the rat closely resemble
the compartmentation of AChE in the flocculus of the rabbit. A C2 zone,
innervated from the rostral pole of the medial accessory olive, is present
39
in the lateral and caudal region of the flocculus of the rat. The dorsal cap
and the ventrolateral outgrowth each project to two discrete zones in the
flocculus of the rat2. The FD and FD' zones receive their climbing fibers
from the ventrolateral outgrowth. They continue in the ventral
paraflocculus, where they are innervated by the ventral leaf of the
principal olive. Two other zones (FE and FE') receive climbing fibers from
the dorsal cap. The FE' zone is found at the rostral pole of the flocculus
and does not extend into the ventral paraflocculus and may correspond to
FC4 of the rabbit flocculus. The FE zone continues for a short distance
into the ventral paraflocculus. where the FD and FD' zones fuse around it.
This fusion can be considered as the equivalent of the fusion of FC ll and
FC3 around FCz in folium p of the rabbit.
The observations of Yamamoto and Shimoyama (1977) on Purkinje
cell zones in the flocculus and folium p of the rabbit are very similar to
our observations on compartments, with respect to their number and
extent. Originally Yamamoto and Shimoyama (1977) distinguished two
pairs of Purkinje cell zones that were retrogradely labeled from the
medial vestibular nucleus and the superior vestibular nucleus.
respectively. The Purkinje cell zones that project to the medial vestibular
nucleus interdigitated with the two zones projecting to the superior
vestibular nucleus. The most rostromedial of these zones projected to the
medial vestibular nucleus. In later papers (Yamamoto, 1978, 1979a, b).
the existence of the most rostral one of the two zones, projecting to the
medial vestibular nucleus was ignored. Purkinje cells in the caudal
flocculus projected to the lateral cerebellar nucleus and those from an
intermediate region of folium p to the nucleus prepositus hypoglossi
(Yamamoto, 1978).
Sato et aL (1982a.b) distinguished three Purkinje cell zones in the
flocculus of the cat: the rostral zone projected to the superior vestibular
2The dorsal cap and the ventrolateral outgrowth are segments of a continuous cell column. Rostrally. this column is continuous with the principal olive. The morphological and functional criteria used to subdivide this cell column are discussed in chapter 2. In the rabbit and cat, the column is subdivided into two subnuclei that comprise the caudal dorsal cap and the rostral dorsal cap with the ventrolateral outgrowth. In the rat. these subnuclei correspond to the dorsal cap and the ventrolateral outgrowth.
40
nucleus. the middle zone projected to the medial vestibular nucleus. and
the caudal zone projected to group y and the parvicellular part of the
lateral. cerebellar nucleus. Three similar. if not identical zones received
projections from the rostral dorsal cap and ventrolateral outgrowth
[rostral and caudal zones). and the caudal dorsal cap [middle zone) [Sa to
et al.,l983). Sato's findings are difficult to reconcile with the data from
the literature and our study on the subdivision of the rabbit flocculus, and
with the more detailed analysis of the olivocerebellar projection to the cat
flocculus by Gerrits and Voogd [1982). Neither Yamamoto [1978) nor
Sato et al. [1982b) reported a zone corresponding to the floccular Cz
zone found in the present study to be related to the posterior interposed
nucleus. Instead, they advocated a projection to the lateral cerebellar
nucleus from Purkinje cells in this region. Their injections of the lateral
cerebellar nucleus, however, may have included the posterior interposed
nucleus. Yamamoto [l979a) reported a projection from the rostral pole of
the MAO to the caudal flocculus tn the rabbit, a projection that is in
accordance with the presence of the C2 zone in this region.
Yamamoto and Shimoyama [1977). Yamamoto [1978,!979a). Gerrits
and Voogd [!982) and Ruigrok et al. [1992) traced their Purkinje cell
and climbing fiber zones from the flocculus into the adjoining part of the
ventral paraflocculus. Gerrits and Voogd [1982) found an extension of the
climbing fiber zones from the dorsal cap, the ventrolateral outgrowth, and
the MAO into their medial extension of the ventral paraflocculus (ME) and
an exclusive projection of the principal olive to the ME and the ventral
paraflocculus. Similar connections from de, vlo. and caudal principAl olive
were described by Yamamoto [1978, 1979a) for folium p of the rabbit
cerebellum and by Ruigrok et al. [1992) for the ventral paraflocculus of
the rat. We also found that compartments Cz and FC 1-FC3 extended into
the white matter of folium p. FC2 and FC3 were only represented in the
rostral pole of folium p. FC4 did not enter folium p.
41
1.4.5 Group y and parvicellular part of the lateral cerebellar nucleus.
Group y can be considered the bed nucleus of the floccular
peduncle. Unfortunately, a definition of this cell group that can be applied
to all species is lacking. Broda! and Pompeiano (1957) first described it
as a subgroup of the vestibular nuclei of the cat. containing small cells.
capping the restiform body dorsally. It was situated between Deiters' nucleus medially and the cochlear nucleus laterally. Scattered cells
connected it with the lateral cerebellar nucleus. Group y was further subdiVided by Gacek (1977;1979) and Highstein and Reisine (1979), both
in the cat. Highstein and Reisine (1979) distinguished a dorsal group y,
located within the floccular peduncle where it arches over the restiform
body (corresponding to the infracerebellar nucleus of Gacek) from more ventrally located, small and tightly packed cells of the ventral group y
(corresponding to group y of Gacek). The two nuclei differed in their
connections. the dorsal group y projecting to the oculomotor nucleus,
while the ventral group y is part of the commissural vestibular system. No
descriptions of group y are available for the rabbit. We distinguished
dorsal and ventral subdivisions of group yin AChE-reacted material of the
rabbit. Ventral group y corresponds to the densely. AChE-stained
parvicellular layer that caps the restiform body. The larger. AChE-positive
cells of dorsal group y are scattered between the fibers of the floccular
peduncle. Epema eta!. (1988), Gerlits eta!. (1989). Epema et a!., (1990)
and Thunnissen ( 1990) included a valiable portion of Lpc in the rabbit in
their group y (sometimes indicating it as the "dorsal group y"). because
the Lpc participated both in the vestibula-oculomotor and commissural
projections. The position of Lpc in rabbit corresponds to that of the
parvicellular part of the lateral cerebellar nucleus of the rat (Korneliussen.
1968; Voogd et a!., 1985; Ruigrok and Voogd, 1990). The border
between the AChE-positive Lpc and the strongly reactive group y is not
sharp and both nuclei merge with the AChE-positive caudal pole of the
superior vestibular nucleus, which was indicated as a separate subnucleus
(group l) by Broda! and Pompeiano (1957) in the cat. In our subdMsion of
the cerebellar nuclei we closely followed Ono and Kato (1938). The main
42
divergence from these authors is the identification of the "pars a" of the
posterior interposed nucleus of Ono and Kato (1938). and also of Van
Rossum (1969), as the dorsolateral protuberance of the fastigial nucleus.
first described by Goodman, Hallett and Welch (1963) in the rat
The explanation for the presence of AChE at the borders of white
matter compartments of the rabbit flocculus and paraflocculus still eludes
us. but as an independent method of reliably subdividing that part of the
cerebellum it has proven useful to correlate the connections.
immunohistochemistry and electrophysiology of the rabbit flocculus (Tan
et aL. 1989; Vander Steen et al., 1991, Chapter 4).
43
44
Chapter 2
Zonal organization of the climbing fiber projection to the flocculus
and nodulus of the rabbit.
2.1 llntroductlon
The flocculus and nodulus are involved in control of motor systems relayed by the vestibular nuclei. Best known is the role of the flocculus in
the adaptive control of eye movements (Robinson, 1976; !to, 1982, 1984;
!to et al., 1972. 1982a, b; Nagao, 1983, 1988; !to 1990). The flocculus
receives a visual input Via climbing fibers from the inferior olive. The
nodulus receives climbing fibers both from visually-dominated and
vestibular subnuclei of the inferior olive. The main output of the flocculus
and the nodulus is directed. through their Purkinje cell axons. to the
vestibular nuclei. Climbing fibers terminating in the flocculus originate from the
dorsal cap (de) and adjoining ventrolateral outgrowth (vlo) of the Inferior
olive (Maekawa and Simpson, 1972. 1973; Alley et al.. 1975; Maekawa
and Takeda, 1976, 1977; Hoddevik and Broda!, 1977; Groenewegen and
Voogd, 1977; Kawamura and Hashikawa. 1979; Walberg et al., 1979;
Yamamoto, 1979a; Broda! and Broda!, 1982; Gerrits and Voogd, !982,
1989: Blanks et al .. 1983; Sato et al .. 1983; Whitworth et al., 1983;
Balaban, 1984; Takeda and Maekawa 1984a. 1989a,b; Langer et al .. 1985;
Bernard, 1987; Blanks. 1990; Ruigrok et al .. 1992) and the rostral pole of
the medial accessory olive (MAO) (Broda!, 1940; Alley et al., 1975;
Hoddevik and Broda!, 1977; Walberg et al., 1979: Yamamoto. 1979a;
Broda! and Broda!, 1982; Gerrits and Voogd, 1982; Blanks et al .. !983:
Langer et al., 1985; Ruigrok et al .. 1992). Climbing fibers to the nodulus
and the adjoining part of the uvula take their origin from the de and vlo,
the subnucleus .13. the dorsomedial cell column (dmcc). rostral and caudal
parts of the MAO and the ventral leaf of the principal olivary nucleus
45
(Broda!, 1940; Alley et al., 1975; Broda!. 1976; Groenewegen and Voogd,
1977; Groenewegen et al., 1979; Kawamura and Hash!kawa, 1979; Voogd
and B!gare. 1980; Broda! and Kawamura, 1980; Broda! and Broda!. 1981,
1982; Whitworth et al., 1983; Furber and Watson, 1983; Eisenman. 19·84;
Sato and Barmack, 1985; Walberg et al.. 1987; Bernard. 1987·. Katayama
and Nisimaru, 1988; Balaban and Henry, 1988; Kanda et al.. 1989; Kana
et al., 1990). Some of the neurons of the de and the vlo distribute their
axons to both the flocculus and the nodulus and uvula (Takeda and
Maekawa. 1984, 1989a,b; Maekawa et al., 1989).
Visual signals are sent to the dc/vlo Via the nucleus of the optic
tract (NOT). the nuclei of the accessory optic system (AOSJ and the visual
tegmental relay zone (VTRZ}. The projection of the optic nerve to the
NOT and the nuclei of the AOS IS mainly crossed. The caudal part of the
de receives afferent projections from the dorsal (DTN) and interstitial
terminal nucleus (ITN, i.e. the interstitial nucleus of the superior fascicle
of the AOS, Giolli et al., 1984, 1985, 1988 and the NOT (Mizuno et al.,
1973, 1974; Takeda and Maekawa, 1976; Holstege and Collew!jn, 1982).
This projection is mainly uncrossed and also involves the group .B (Holstege and Collew!jn, 1982). The rostral part of the de and the vlo
receive afferent projections from the medial terminal nucleus of the
accessory optic system (MTN) and the VTRZ (Maekawa and Takeda,
1977, 1979: G!olli et al., 1984, 1985; Simpson et al.. 1988). The major
projection from these nuclei is ipsilateral and derived from the VTRZ. The VTRZ does not receive direct afferents from the optic tract, but it
receives a crossed projection from the MTN, passing through the
posterior commissure (Giolli et a!., 1984). Physiologically, the dc/vlo
complex can also be subdlvlded Into caudal and rostral parts on the basis
of laterality of retinal signals (Maekawa and Takeda, 1976, 1977; Ito et
al., 1978; Takeda and Maekawa, 1980), their connectivity With different
vestibula-ocular reflexes (Ito et al.. 1978, l982b) and their preferred axis
in responses to rotating visual patterns (Simpson and Alley 1974; Simpson et al., 1981; Leonard et al., 1988).
The de also receives a projection from the pontine paramedian
reticular formation (Gerrits and Voogd, 1986). Afferents from the
46
vestibular nuclei terminate in different subnuclei of the olive including
subnucleus b of the caudal MAO, subnucleus j3 and the dmcc (Saint-Cyr
and Courville, !979; Martin et al., !980; Gerrits et al.. !985; Nelson et
al., !986; Nelson and Mugnaini, !989; Kaufman et al., 1991). The nucleus
prepositus hypoglossi projects both to Visual and vestibular subnuclei of
the inferior olive (Gerrits et al., 1985, McCrea and Baker, 1985). This
projection is partially GABAergic (De Zeeuw et al., 1993), partially
cholinergic (Barmack et al.. 1993).
A zonal pattern in the olivocerebellar projection to the flocculus,
consisting of a central zone from the caudal de and two flanking zones
from the rostral de and the vlo. was first demonstrated with
autoradiography of axonal transport of tritiated leucine by Groenewegen
and Voogd (1977) in the cat. These results were confirmed in retrograde
tracing studies In rabbit (Yamamoto, 1979a) and cat (Kawamura and
Hashikawa, 1979; Sato et al., 1983). The three climbing fiber zones seem
to coincide With three Purkinje cell zones that were delineated in the
rabbit (Yamamoto and Shimoyama, 1977) and the cat (Sato et al., 1983)
on the basis of their projection to different vestibular and/or cerebellar
nuclei. A more complicated pattern with seven climbing fiber zones was
described by Gerrits and Voogd (1982) for the olivocerebellar projection
to the flocculus of the cat. They identified the caudalmost zone of the
flocculus as an extension of the C2 zone, which receives its climbing
fibers from the rostral pole of the MAO. Gerrits and Voogd ( 1982)
stressed the extension of some of the floccular zones (including the C2
zone) into the adjoining part of the ventral paraflocculus (the medial
extension of the caudal ventral paraflocculus or ME). The ME differs from
the flocculus in rece!vlng an additional projection from the principal
olive. A projection of the principal olive to folium p, which is the
equivalent in the rabbit of the ME, was also described by Yamamoto (1979a).
A zonal arrangement in the olivocerebellar projection to the
flocculus and the adjacent ventral paraflocculus is also present in the rat
(Ruigrok et a!., 1992). Five climbing fiber zones were distinguished,
including a C2 zone in the caudal flocculus and two pairs of alternating
47
zones, innervated by the caudal de and the rostral de With the vlo
respectively. All but the most medial. which is innervated by the caudal
de, extended into the paraflocculus. Parasagittal zonal patterns in the
olivonodular projection were found by several authors; the most detailed
pattern was presented by Katayama and Nisimaru (1988) and Balaban and
Henry (1988). both in the rabbit.
It is clear that not all of these studies agree on the location and
number of climbing fiber zones in the flocculus. Still. a pattern has
emerged of discrete zones (Ito. 1984) or modules (Voogd and Bigare.
1980) consisting of subsets of Purkinje cell receiving specific climbing
fiber input and projecting to distinctive parts of the vestibular and
cerebellar nuclei. These modules may constitute the anatomical circuitry
used by the vestibulo-ocular (Ito et al., 1977: Ito et al., 1982b: Sato and
Kawasaki, 1990b) and eye movement related zones in the flocculus to
control eye muscle activity (Dufosse et al .. 1977; Balaban et al., 1984:
Nagao eta!., 1985; Sato and Kawasaki, 1984, 1990a, 1991). An intrinsic
framework for revealing these zones, such as the one revealed by the
AChE-positive raphes that subdivide the white matter of the flocculus
(Hess and Voogd, 1986: Voogd et al., 1987), has never been used to
correlate anatomical and physiological data. Such a framework became
available with our recent subdivision of the white matter in AChE-stained
sections of the rabbit flocculus into five compartments (see Chapter 1).
This study addresses the following questions:
a) Do the afferent climbing fibers from different parts of the dc/vlo
complex travel in specific white matter compartments in the flocculus
and nodulus ?
b) What is the relationship between the white matter compartments and
the climbing fibe'r termination zones ?
48
2.2 Materials and methods
In 20 pigmented Dutch belted rabbits small quantities (5-15 nll of a
7% solution of horseradish peroxidase coupled to wheat germ agglutinin
(WGA-HRP, Sigma type VI) were injected with air pressure through
micropipettes, placed in the inferior olive. A dorsal approach through the
fourth ventricle avoided the cerebellum. In nine cases optimal sites for
injection were located by electrophysiological recording. using the
method and the criteria for the responses of dc/vlo neurons to rotatory
stimuli of Leonard et al. (1988). Recordings and injections were made
through the same glass pipette filled with the WGA-HRP solution.
Neurons in the caudal de can be recognized by their preference for
rotation of the visual surround about the vertical axis, rostral dcjvlo
neurons prefer rotation about specific horizonal axes. Eye dominance can
also be used to distinguish caudal de and rostral dc/vlo responses during
Visual stimulation. Flash stimuli to the ipsilateral and contralateral eye
were used to determine the relative rostrocaudal position of the tip of the
injection pipette. Flash potentials from caudal de neurons were optimal
when the stimulus was delivered to the contralateral eye. Rostral de
neurons responded optimally with flash stimulus to the ipsilateral eye,
while vlo neurons and the most rostral neurons in the rostral de hardly
showed any response to flash stimulation.
After a survival time of 36-48 hours the rabbits were deeply
anaesthetized and perfused transcardially with 1 L of saline at room
temperature, followed by 2 L of a solution containing 1-3%
paraformaldehyde and 0.5-1.25% glutaraldehyde in 0.1 M phosphate
buffer (pH 7.2-7.4) at room temperature. and finally With 1 L of 10%
sucrose tn 0.1 M phosphate buffer (pH 7.2-7.4) at 4'C. The brainstem and
cerebellum were· removed, rinsed in 10% sucrose phosphate buffer, and
embedded in a solution of 10% gelatin and 10% sucrose. The embedded
tissue was stored overnight in 30% sucrose in 0.1 M phosphate buffer (pH
7.2-7.4) and then cut into 40 J.Lm transverse sections. Adjacent serial
sections were processed with diaminobenzidine as a chromogen in the
histochemical reaction for horseradish peroxidase (one series, DAB,
49
Graham and Karnovsky, 1966) or tetramethyl benzidine (two series, TMB.
Gibson et al., 1984) as the chromogen and acetylcholinesterase (one
series, AChE). A standard thiocholine method (Geneser-Jensen and
Blackstad, 1971) was used with acetylthiocholine iodide as a substrate
and ethopropazine as a non-specific acetylcholinesterase blocker (Hess
and Voogd, 1986). The extent of the injection site was assessed in DAB
reacted sections With light microscopy. The climbing fibers in the white
matter and molecular layer were plotted from TMB reacted sections and
their position was compared to the AChE-stained raphes in adjacent
sections. The serial DAB- and TMB-processed sections of one case with
an injection of WGA-HRP in the rostral MAO (B 2202) was put at our
disposal by Dr. C. Yeo and J. van Ham, MD from the neurobiology group of
the Department of Anatomy, University College, London U.K.
In three rabbits (C 914, C 1134, C 1135) tritiated leucine was
injected in the caudal inferior olive. A stock solution of L(4,5-3H)-leucine
(specific activity about 140 Ci/mM, concentration 1 fLCi/fil; Radiochemical
Centre. Amersham, UK) was evaporated to dryness under a gentle flow of
nitrogen at 40°C and redissolved in saline to a final concentration of 75
fLCi/fil. Injections were performed with a l Ill Hamilton syringe with a 25
gauge needle. To reduce unwanted spread of tritiated leucine delivery
started 5 minutes after placing the needle in position. The injections
were made at a rate of 0.1 1-1-l per 10 minutes. After the injection the
needle was left in place for an additional 15 minutes. FolloWing a survival
time of 4-7 days the animal was transcardially perfused under deep
anaesthesia with 0.5 L of saline and 2 L of 4o/o formaldehyde. The
brainstem and cerebellum were removed and embedded in 1 Oo/o gelatin
and transversely sectioned at 30 IJ.ID on a freezing microtome. Sections
were mounted on slides prepared With a chrome-alum gelatin solution.
defatted in xylene, dipped in IIford G5 emulsion, and exposed for 6
weeks at 4°C. The sections then were developed in Kodak Dl9 at 16-l8°C
for 4 minutes. rinsed in distilled water, fixed in 24o/o sodium thiosulphate
for 8 minutes. and counterstained With cresyl violet. Criteria previously
described by Groenewegen and Voogd (1977) were used to delineate the
effective injection site. All injection sites are illustrated in selected
50
sections and in graphical reconstructions of the inferior olive, prepared
according to Broda! (1940) and Gerrits and Voogd (1982). The caudal de
and the rostral de are separated by a narrow neck (Leonard et al.. 1988).
The border between the rostral de and the vlo is arbitrary. The vlo is
continuous with the dorsal leaf of the principal olive.
fig. 2.1 Diagram illustrating the unfolding of the Purklnje cell layer used to prepare maps of the unfolded surface of the flocculus. The Purkinje cell layer A-B is straightened: the position of labeled climbing fibers (in this case FZn) is indicated. Sections are aligned using the distance between the midline (M) and the most ventromedial point of the Purkinje cell layer {B). Four sections (A-B to A"' to B"') are reconstructed in C. The distance between the reconstructed sections (t) is obtained by multiplying the distance between the sections by the magnification factor.
The labeling of the climbing fibers was represented in diagrams of
the unfolded Purklnje cell layer of the flocculus and the adjacent ventral
paraflocculus. The length of the unfolded Purkinje cell layer in a single
transverse section was indicated as a straight line (fig 2.1 A. B). Segments
of this line representing the portions of the molecular layer containing
the labeled climbing fibers were demarcated. The sections were aligned
using the distance between the midline (M) and the most ventromedial
point of the floccular Purkinje cell layer (B in fig. 2.1). The distance
51
between the reconstructed sections {t) corresponds to the distance
between the sections multiplied by the magnification factor. Finally, the
points on the lines representing the borders between the folia and the
climbing fiber zones were interconnected (fig. 2.1).
The folial rosette of the flocculus usually consists of four or five folia
which fuse caudally. They are indicated from ventral to dorsal as folium m
and fl-f4. The cortices of the flocculus and the adjoining folium of the
ventral paraflocculus (folium p) are continuous in the bottom of the
posterolateral fissure (Yamamoto and Shimoyama, 1977). but sometimes
interrupted by surfacing white matter (figs. 1.2, 1.3 D). The
compartments of the white matter of the flocculus are numbered from
lateral to medial as C2 and FC1-FC4. Raphes are indicated with the
numbers of the compartments they separate (e.g. l/2). Climbing fiber
zones are indicated With the prefix FZ (floccular zone) and the Roman
numeral of the corresponding compartment. Compartments in
AChE-stained sections containing lobule X, and the corresponding
climbing fiber zones, are indicated as XC1 -XC4 and XZr-XZv respectively.
The distribution of retrograde labeling In the AOS nuclei and VTRZ in
experiments with injections of WGA-HRP into the inferior olive was
determined and compared with the rostrocaudal extent of the injection
site in the de and the vlo (fig. 2.8).
2.3 Results
2.3.1 Projections of the caudal dorsal cap (caudal dd.
In slx rabbits the targeted injection site was in the caudal half of the
dorsal cap (caudal de). In four cases (K 324, K 369, K 426, C 914) the
injection site did not extend beyond the narrow neck separating the
caudal de from the rostral dorsal cap (rostral de). In two other cases
(K 314 and K 325) the Injection site included the caudal part of the
rostral de. One experiment out of each of these two groups will be
described In some detail.
52
The injection site in case K 369 (fig. 2.2 B) includes in addition to
the caudal de. the medial part of the caudal MAO and subnucleus .B. Retrogradely labeled neurons in the AOS are limited to the nucleus of the optic tract (NOT). dorsal (DTN) and interstitial (!TN) terminal nuclei (fig.
2.8 A). Climbing fibers from the caudal de cross the midline, pass lateral
to the spinal tract of the trigeminal nerve and accumulate along the
medial border of the restiform body and more rostrally along the dorsal
border (fig. 2.3 J). Many labeled fibers enter the cerebellum along the
superior cerebellar peduncle. Within the cerebellum they pass rostral and
dorsal to and through the anterior interposed nucleus to be distributed to
the white matter of the vermis and the medial part of the hemisphere (fig. 2.3 d-m). Labeled fibers from the caudal de enter the floccular white
matter medially. Some of these fibers arch through or around group y and
the ventromedial extension of the lateral cerebellar nucleus (Lpc). Within
the flocculus they separate into two streams. One stream passes through
the medial compartment FC3 to accumulate in compartment FCz (figs.
2.3 h: 2.4 H). Within FC2 the fibers are densely packed, especially
medially. Their distribution in FCz is complementary to the AChE
staining. which is more concentrated in the lateral part of compartment
(fig. 2.4 H). From this compartment labeled fibers enter the granular
layer to terminate as climbing fibers in a continuous strip in the
molecular layer of folia fl-4, and also the rostral pole of folium p (fig. 2.2
A; 2.3 b-g; 2.4 A-D). The border between labeled and unlabeled parts of
the cortical layers is sharp. Climbing fiber zone FZn. composed of the
terminals of olivocerebellar fibers from FCz. gradually shifts laterally in
more rostral sections (fig. 2.3 a-!; 2.4).
Labeled fibers of the second fiber stream accumulate at the border
of the white matter of the flocculus with the restiform body and the
brachium pontis along the AChE-positive cell strands in this region (fig.
2.3 f-h). Where the raphe between FC3 and FC4 can be clearly
distinguished in the rostral flocculus they become located in FC4, where
they travel in a rostrodorsal direction (fig. 2.3 a-f: 2.4 A-C). They
terminate as the climbing fiber zone FZIV in the medial parts of f3 and f4
53
A - =lmm
[ - =lmm
v.;_l---~ dl
__ ('
D
·--· -- -- ------._
K .114 cdc~ - -~· --~·--\'A
54 ~
(fig. 2.2 A). FZ!V does not extend into folium p. FZn and FZ!V are
separated from each other by an empty strip of molecular layer
corresponding to FZm. which is innervated from FC3 (see next
paragraph). Labeled climbing fibers from this injection site terminate in
isolated patches in caudal folium p without a clear zonal pattern. This
labeling cannot be associated with a specific compartment or climbing
fiber zone (fig. 2.2 A).
Labeled olivocerebellar fibers in FC2 can be traced caudally in the
roof of the lateral recess. ventral to the lateral cerebellar nucleus. They
continue their course in the roof of the fourth ventricle, near the
attachment of the posterior medullary velum to enter the nodulus at its
lateral side (fig. 2.3 i-k). Here they terminate in two climbing fiber strips.
one In the medial and one in the lateral part of this lobule.
Four compartments can be distinguished in the white matter of the
nodulus in acljacent AChE-stained sections (fig. 2.3 k-m; 2. 7; 2.11 k-1;
2.12 m-n). Rather wide compartments XC! and XCz are present in the
medial two thirds of the lobule. A distinct AChE-positive raphe delineates
XC3, which is only present in the dorsal white matter of the nodulus (fig.
2. 7 A, C). It corresponds to a similar, but wider compartment in the
lateral white matter of the uvula. Compartment XC4 is narrow and
occupies the lateral white matter of the nodulus. Labeled fibers are
located in XC4 and XC! (figs. 2.6 a; 2.7 G, H). Those contained in XC4
terminate in two concentrations of climbing fibers on the ventrolateral
aspect of the nodulus and in a narrow single strip on the dorsal aspect of
this lobule. Labeled fibers in XC 1 innervate the XZ1 zone, flanking the
fig. 2.2 Drawings of the distribution of labeled climbing fibers in the unfolded flocculus and ventral paraflocculus and the injection sites in the caudal dorsal cap. The presence of labeled climbing fibers in the sections used to prepare the unfolded flocculus is indicated by horizontal lines. Climbing fibers of the FZIV zone are indicated with large dots, those belonging to FZn with small dots. Climbing fibers In caudal folium p cannot be allocated to these zones. Injection sites are indicated in a diagram of the unfolded inferior olive and in assorted transverse sections. A and B: case K 369; C and D: case K 314.
55
midline. This zone continues from the ventral side of the lobule X onto its
dorsal side and farther into lobule IX. A rostrolateral triangular area.
indicated with a question mark in fig. 2.6 a, is clearly separated from the
rest of XZ1.
Other fibers head for the midline dorsal to the cerebellar nuclei and
constitute a parasagittal band of labeled fibers in the molecular layer of
the vermis close to the midline. They correspond to the A zone of
Groenewegen and Voogd (1977). Labeled olivocerebellar fibers enter the
white matter of the paramedian, ansiform, and simple lobules from the
region of the dorsolateral protuberance of the fastigial nucleus. They
terminate as a continuous climbing fiber zone in the medial part of the
molecular layer of these lobules (fig. 2.3 h-k) and correspond to the
lateral A zone of Buisseret-Delmas (1988). The medial A zone and the
lateral A are innervated from the subnuclei b and c of the caudal l'viAO,
which are included in the injection site in this case.
Retrogradely labeled cells are present mainly in the medial part of
the fastigial nucleus. A few labeled cells are located in the lateral
cerebellar nucleus. Retrogradely labeled cells in the vestibular complex
are found in the caudal descending vestibular nucleus (DV) bilaterally and
in the ventral part of the ipsilateral medial magnocellular vestibular
nucleus (MVmc). Very few labeled neurons are present in the SV, none
are located in group y (fig. 2.3 d-k).
The injection site in case K 314 includes the caudal MAO. caudal
DAO, subnucleus J3. and the caudal de, and a bit of the caudal part of the
rostral de (fig. 2.2 C). The distrtbutlon of retrogradely labeled neurons in
the AOS nuclei is similar to cases K 369 and K 426, but in addition a
fig. 2.3 DraWings of AChE-stained sections through the flocculus and the caudal vermis (left columns) and of adjacent sections incubated for HRP through the flocculus and the cerebellum (right columns) in case K 369. Note the presence of labeled olivocerebellar fibers in the FC2 and FC4 compartments of the flocculus and the XC 1 and XC4 compartments of the nodulus and the termination of these fibers in corresponding zones. In addition. olivocerebellar fibers terminate in the A zone of the anterior vermis (A) and in the lateral A zone (dlp) of Buisseret-Delmas (1988).
56
57
fig. 2.4 Darkfield photomicrographs of four sections through the flocculus in case K 369 (A. rostral; G, caudal) and brtghtfield photomicrographs of adjacent AChE-stained sections (B-H).
58
...... H ..
59
60
small number of weakly labeled neurons was observed in the VTRZ.
Labeled fibers are located in compartments F'C2 and FC4 and terminate in
the corresponding climbing fiber zones FZn and FZIV (f!g. 2.5; 2.10 A-B.
D). In this case a third fiber stream is present. It separates from the
second fiber stream in the caudal flocculus and travels in FC1 in a lateral
direction towards folium f2 and f3 to terminate in the most lateral parts
of the molecular layer of these folia (figs. 2.3. C; 2.5 d-g, arrows). As in
case K 369 labeled cl!mb!ng fibers are also found in the caudal part of
folium p. The labeling appears as patches in the molecular layer and
cannot be associated With any of the floccular zones or compartments (fig.
2.5 i-k).
The distribution of labeling in the XC 1 and XC4 compartments and
the XZI and XZrv zones of the nodulus is similar to the preVious case {figs.
2.5 l; 2.6 b). Climbing fiber labeling was present in the A and B zones of
the anterior vermis and in the lateral A zone of Buisseret-Delmas (1988)
in the hemisphere of the posterior lobe.
The distribution of tritiated leucine-labeled climbing fibers in case
C 914, with an injection in the caudal de and the subnucleus J) (fig. 2.14
A) is similar to the preVious two cases (fig. 2.6 C; 2.7 H).
fig. 2.5 Drawings of adjacent AChE-stained and HRP-reacted transverse sections through brain stem and cerebellum in case K 314. Note the presence of labeled fibers in compartments FC2 and FC4, which terminate as climbing fibers in corresponding zones in the molecular layer of the flocculus. Some labeled fibers are present in compartment FC1, terminating in FZI (arrowheads in d-f). Labeling in lobule X is present in compartments XC 1 and XC4. For symbols indicating compartments in AChE-stained sections see fig. 2.3.
61
apex
I -~ @ .:,;..
dorsal
~ K369 caudal DC+~
@
dorsal .. K 328 rostral DC + VLO
I '{1],
""'" :48 ..
.
K 314 caudal DC+~
I
I
K397~~1lJ rostral DC + VLO and DM(C
®
62
C914 caudal DC+ ~
[ 1135 VLO + rostral MAO and DMCC
CD
i
apex
I dorsal
1 c 1134 entire caudal inferior olive; rostral MAO and DMCC spored
®
B 2202 rostral MAO + OMCC
1--; 1mm
® CJ)
wO,ternot"'
(OC~O<
fig. 2.6 Maps of the climbing fiber distribution on the ventral and dorsal surface of lobule X. The presence of labeled climbing fibers in the sections used to prepare the maps is indicated by lines. Climbing fiber zones are numbered I-V and indicated by dots. The intermediate, rostral dc/vlo-innervated strip in XZIV is indicated as XZJv•. Arabic numerals refer to the white matter compartments XC1-XC4 illustrated in the bottom diagram in fig. 2.6 1. The ventromedial compartment indicated with a questionmark was labeled after injections including the group _13, the caudal and rostral dorsal cap {a -d) and in one of the cases involving the ventrolateral outgrowth {e). A summarizing diagram of the compartments and the climbing fiber zones of the nodulus is shown under i. The white matter compartment corresponding to xzv cannot be distinguished in AChE-stained sections and is not indicated in i.
63
64
2.3.2 The projection of the rostral dorsal cap and ventrolateral outgrowth
(rostral de/ ulo}.
In five rabbits (K 327. K 328. K 333, K 397 and C 1135) the
injection site involved the rostral dc/vlo. Three of these cases {K 328, K
397 and C 1135) will be presented.
The injection site in case K 328 is limited to the rostral de and vlo
(fig. 2.9 B). Retrogradely labeled neurons in the AOS nuclei are only
present in the VTRZ (fig. 2.8 B). Labeled olivocerebellar fibers, located
along the dorsolateral border of the contralateral restiform body, enter
the floccular peduncle, where they ramify among the cells of group y (fig.
2.11 g-i). Next they occupy compartment FC3 and, arching over FC2, also
distribute to compartment FC1 (Fig. 2.11 f-g). They can be traced
through FC1 and FC3 to terminate as climbing fibers in the FZ1 and FZm
zones (fig. 2.9 A). FZI occupies a lateral position, but remains separated
from the lateral margin of the flocculus by an unlabeled stretch of
molecular layer, corresponding to the C2 zone (see below), which
broadens more caudally. The C2 compartment of the flocculus also does
not contain any labeled fibers (fig. 2.9 A; 2.11 e-g). FZm is separated
from FZt by an empty strip, corresponding to FZn (fig. 2.11 a-h). FZr and
FZm extend into folium p, enclosing the dorsal tip of FZn (fig. 2.12 e,
Fig. 2. 7 Photomicrographs of AChE-stained sections and darkfield exposures of sections containing labeled fibers in lobules IX and X. A.B.F: case K 397. A,B: AChE. The raphes bordering compartment 3. which is present in dorsal lobule X and occupies the lateral white matter of lobule IX and the raphe between the compartments XC 1 and XC2 are indicated With small, white arrows in A and E. The compartments are indicated with arabic numerals. WGA-HRP labeled fibers from the rostral de and the vlo are present in XZn and XC2. Labeled climbing fibers from the dmcc are present in XC3 and terminate laterally in lobule IX (arrow). C.D: case K 328. WGA-HRP labeled fibers from rostral de and vlo are present in compartments XC2 and XC4 and corresponding zones. F.G: case K 369. WGA-HRP labeled fibers after an injection in caudal de and subnucleusJ3 are present in XCI and XC4. H: Case C 914, with an injection of tritiated leucine in caudal de and subnucleus .13. Distribution as in previous case K 369.
65
;c
--,,.," NOT.j
·~·
@ K369
; ITN I ,,
(,
/'Pc
fig. 2.8 DraWings of transverse sections through the mesencephalon in cases K 369 (A) with an injection involving the cdc and K 328 (B) with an injection involving the rdc/vlo. Black dots indicate retrogradely labeled cells. The VTRZ is labeled in case K 328; the nucleus of the optic tract (NOT), the dorsal and interstitial nuclei of the accessory optic tract (DTN, !TN) in case K 369.
66
ventral
B
PFLv c -=lmm
fig. 2.9 Diagrams of the distribution of climbing fibers in the unfolded flocculus and ventral paraflocculus and diagrams of the inferior olive showing injection sites including the rostral dorsal cap and ventrolateral outgrowth, in cases K 328 (A,B) and K 397 (C,D). Labeled climbing fibers in sections used to prepare the unfolded flocculus are indicated with horizontal lines. Labeling in FZr is Indicated with coarse hatching and in FZm with fine hatching.
67
fig. 2.1 0 Darkfield photomicrographs of corresponding TMB-processed sections through the flocculus inK 314 (A) with an injection in the caudal dorsal cap and K 327 (C} With an injection in the rostral dorsal cap. Raphes and white matter compartments are depicted in B and D. Note that the labeled parts of the white matter and associated floccular cortex in K 314 (FCz. FC4 and FZn) are not labeled in K 327 which shows prominent labeling in FC J, FC3 and FZ1 and FZm.
arrow). Patchy climbing fiber labeling is present in the molecular layer of
the central part of folium p (fig. 2.!! f-i). and it cannot be correlated to
any of the FC or FZ.
Labeling in compartment FC3 can be traced caudally, ventral to the
lateral cerebellar nucleus and medially along the attachment of the roof of
the fourth ventricle (fig. 2.11 j). These fibers subsequently enter lobule X.
Here they occupy compartments XC2 and XC4. which can be delineated in
adjacent AChE stained sections (fig. 2.11 k-1). Fibers from XCz terminate
68
fig. 2.11 DraWings of adjacent AChE and HRP-reacted transverse sections through brain stem and cerebellum in case K 328. Note the presence of labeled fibers in FC 1 and FC3 and FZ1 and FZm in the flocculus and in compartments and zones XZ1 and xzw of lobule X. For symbols indicating compartments in AChE-stained sections see flg. 2.3.
69
as climbing fibers in the intermediate part of XZu. ·which is continuous in
the bottom of the posterolateral fissure with a similar zone in lobule IX
(fig. 2.6 d; 2.7 D). Labeled fibers in compartment FC4 terminate in a
narrow lateral strip on the ventral and dorsal aspect of this lobule. The
medial compartment XC1 and the medial cortex do not contain labeled
fibers.
A heavy concentration of retrogradely labeled cells is present in the
Lpc (fig. 2.11 h-i), a few labeled cells are found in the fastigial nucleus.
Neurons of dorsal group y, lying Within the floccular peduncle or ventral
group y, capping the restiform body are not labeled. Labeled fibers pass
medially from the Lpc in the rostral wall of the lateral recess and the
lateral angle of the fourth ventricle (no, fig. 2.11 h). They ascend in a
bundle located ventral and ventrolateral to the superior cerebellar
peduncle (fig. 2.11 b-f) which enters the decussation of the superior
cerebellar peduncle. Their localization corresponds to the nucleo-o!ivary
fibers (Legendre and Courville, 1987). which were retrogradely labeled in
this experiment.
The injection site in K 397 is larger than in K 328 and in addition
to the rostral de and vlo, it also involves the dmcc and the rostromedial
parts of the DAO and MAO. except for their rostralmost poles !fig. 2.9 D).
It also extends into the contralateral olive. Retrogradely labeled cells are
present in the VTRZ. but they were absent from the NOT, DTN. and ITN
(fig. 2.8 B). Hardly any labeled cells were found in the MTN and LTN.
Labeled fibers enter the flocculus via the floccular peduncle (fig. 2.12 i-1),
pass through group y and the anterior interposed and lateral cerebellar
nuclei. The distribution of labeled olivocerebellar fibers in the flocculus is
very similar to the preVious case (K 328). The labeled climbing fibers of
FZ1 and FZm merge in the rostral molecular layer of folium p.
surrounding a small unlabeled area, corresponding to the FZn zone (.Fig.
2.12 e, arrow). Caudally patchy labeling extends into the central part of
folium p and into other parts of the ventral paraflocculus (fig. 2.12 g-j).
The labeling in lobule X also resembles the previous case. In addition,
some labeled fibers are present in compartment XC3 (fig. 2.12 n) and
70
they terminate in XZm in the lateral half of the dorsal molecular layer of
lobule X. adjacent to XZn and the ventral molecular layer of lobule IX (figs.
2.6 e; 2.12 m-n). Labeling in the hemisphere of the anterior and
posterior lobes is fairly extensive. In the white matter of the caudal
vermis labeled olivocerebellar fibers surround an empty compartment
(fig. 2.12 1. asterisk), which contained the fibers from the lateral A-zone
in experiments with injections involVing the caudal and medial MAO (figs. 2.3 k; 2.5 k-1).
Extensive retrograde labeling is present among neurons of the
cerebellar nuclei. including the Lpc (fig. 2.12 i-1; 2.13 D). The neurons of
dorsal group y located between the fibers of the floccular peduncle and of
the ventral group y located ventral to the peduncle and dorsal to the
restiform body are not labeled. Some labeled cells are located in the
medial and descending vestibular nuclei. The nucleo-olivary fiber bundle
situated in the lateral recess of the fourth ventricle and ventral to the
superior cerebellar peduncle is heavily labeled (no. fig. 2.12 d-h; 2.13).
In one case (C 1135, fig. 2.14 B) an injection of tritiated leucine
was made in the rostral de and vlo, but it also extended widely into other
parts of the inferior olive, including the rostral pole of the MAO, which is
thought to contain neurons projecting to the Cz zone, and the dmcc. In
the rostral dc/vlo the injection site was situated somewhat more rostrally
than in the preVious two cases (K 328 and K 397). The distribution of
fibers in the floccular cortex is similar to them except for the presence of
labeled climbing fibers in the cortex lateral and caudal to FZ!,
corresponding to the C2 zone, which was unlabeled in K 328 and K 397.
In C 1135 labeling in the C2 zone continues into the medial part of the
ventral paraflocculus and the dorsomedial and lateral part of the dorsal
paraflocculus. Some labeled fibers are present in the fastigial and
posterior interposed nucleus. Most are located in and dorsal to the
anterior interposed and the medial part of the lateral cerebellar nucleus.
The white matter of the nodulus contains bundles of labeled fibers
in an intermediate and a lateral position. They terminate as climbing
fibers in a central zone corresponding in position to the XZn zone of the
71
72
fig. 2.12 DraWings of adjacent AChE and HRP-reacted transverse sections through brain stem and cerebellum in case K 397. Asterisk (in I) indicates the position of the lateral A zone and the corresponding compartment of Buisseret-Delmas (1988}, which was labeled in K 369/314 (figs 2.3 and 2.9). Arrow in e indicates extension of non-labeled FZn in folium p. Labeled fibers occupy the compartments FC 1 and FC3 and terminate in the corresponding zones. In the nodulus (m-n) labeling is present in XC1. XC3, and XC4 and terminates in the corresponding zones. For symbols indicating compartments in AChE-stained sections see fig. 2.3
previous cases. In addition an extremely lateral strip of molecular layer
surrounding the lateral pole of the lobule (XZv} and a strip on the dorsal
surface of the lobule (XZm) contain labeled climbing fibers. These fibers
presumably take their origin from the rostral MAO and/or the dmcc.
Regions of the molecular layer corresponding to XZ1 and XZrv remain
virtually empty (figs. 2.6 f; 2.15 A).
Retrograde labeling of cells in the cerebellar nuclei does not occur
With injections of tritiated leucine in the inferior olive. No labeled fibers
are observed in the position of the nucleo-olivary tract.
73
74
A
d' "~mcc''
;
ll C'il-1
fig. 2.14 Diagrams of the injection sites in the inferior olive of cases C 914 and C 1135.
2.3.3 The projection of the rostral pole of the medial accessory olive (MAO)
and the dorsomedial cell column (dmcc).
In one case a small injection of WGA-HRP was confined to the
rostromedial part of the MAO and the dmcc (B 2202. fig. 2.16 D). In
another rabbit large parts of the inferior olive, including the caudal de,
the rostral de and the vlo, were injected with tritiated leucine, but the
rostral pole of the MAO and the dmcc were spared (C 1134. fig. 2.16 B).
labeled climbing fibers in case B 2202 (fig. 2.16 C; 2.17) travel
through the ventromedial part of the restiform body to enter the
cerebellum rather caudally. They curve medially towards the midline over
the anterior interposed nucleus and accumulate in the posterior
interposed nucleus. From this fiber bundle labeled fibers detach to enter
the stalk of the paraflocculus. They terminate in a distinct C2 zone
extending over the dorsal and the ventral paraflocculus and the caudal
part of the flocculus (fig. 2.17 b-h). In the white matter of the flocculus
fig. 2.13 Brightfield (A,C) and darkfield (B,D) photomicrographs of transverse sections through brains tern and flocculus in K 397. A,B: anterograde labeling is present in compartments FC 1 and FC3 and retrograde labeling in the nucleo-olivary pathway (no). C,D: Fibers of the nucleo-olivary pathway (no) collect at this more caudal level near the lateral angle of the fourth ventricle. The floccular peduncle (pfl does not contain labeled fibers.
75
and the paraflocculus these fibers occupy a region corresponding to the
C2 compartment. Other labeled olivocerebellar fibers enter the caudal
vermis, where they terminate in dorsal lobule X and ventral lobule IX in a
position corresponding to the XZm zone (fig. 2.17 f-h). Retrogradely
labeled cells are found in the posterior interposed nucleus and a small
cell group ventral to it. A few retrogradely labeled fibers were observed in
the nucleo-olivary tract in the lateral angle of the fourth ventride (fig.
2.17 a-c).
In case C 1134 the injection with tritiated leucine involved the
caudal de, rostral de and vlo (fig. 2.16 B). It extended into the caudal half
of the MAO and DAO and subnucleusj3, but spared the rostral pole of the
MAO, which contains the neurons projecting to the C2 zone. Dense
labeling was present over all parts of the white matter and the molecular
layer of the flocculus except for a small area in the caudal pole, where all
fig. 2.15 Darkfield photographs of autoradiograms of sections through the lobules IX and X. A} C 1135. This injection involved the rdc, the vlo and the rostral MAO and labeled fibers in compartments XC2 and XC3 and the corresponding zones and in the extreme lateral pole of lobule X (XZv). B) C 1134. This large injection spared the rostral MAO and the dmcc. No labeling is observed in compartment XC3 zones XZm and XZv.
76
folia taper into a single folium (fig. 2.16 A). This area approximately
corresponds with the lateral unlabeled area seen in the caudal flocculus of
rostral dc/vlo injected cases and the labeled Cz compartment in case B
2202. The labeling in the white matter of case C 1134 was characterized
by the relatively sparse labeling at the borders between the
compartments. These areas of sparse labeling seem to coincide with
AChE-positive raphes (fig. 2.18). Labeling In the nodulus spared the XC3
compartment and the XZni zone, and the molecular layer covering the
lateral pole of the lobule, corresponding to XZv (fig. 2.15 B). Labeled
olivocerebellar fibers are also found in the fastlgial nucleus, the medial
part of the vermis and paramedian lobule. the dorsal part of the simple
lobule, and in several small bands in the ventral paraflocculus and the
ventral part of the dorsal paraflocculus. This large injection of
anterograde tracer caused no labeling in the nucleo-olivary tract (fig.
2.18).
77
c 1134 rostra! fl
ventral
A -"lmm
PO
c 1134
B
B 2202
fp _./~
( PFLv
-"lmm
78
2.4 Conclusions
The olivocerebellar projection to the flocculus and nodulus is
summarized in the diagrams of figs. 2.19 and 2.6 i.
1. Labeled olivocerebellar fibers in the flocculus from injections
including the caudal de (i.e. the region caudal to the constricted portion
of this subnucleus. cases K 369. K 314. K 426 and C 914) become located
in compartments FC2 and FC4 of the white matter of the flocculus, which
are delineated by AChE-positive raphes in adjacent AChE-stained
sections. They terminate in the corresponding cortical zones FZn and
FZrv (fig. 2.19). Injections involVing the rostral de and vlo result in
labeling fibers in compartments FC1 and FC3, which terminate in FZy and
FZny. These zones extend into folium p and merge around the rostral tip
of FZn (fig. 2.19). Some labeling in compartment FC1 and the FZr is
observed in K 314 where the caudal de injection site encroaches on the
most caudal part of the rostral de. The C2 compartment is located
laterally and caudally in the white matter of the flocculus. It is continuous
with a compartment in the medial white matter of the paraflocculus and
contains labeled fibers that terminate in the corresponding C2 zone after
injections including the rostral pole of the MAO (C 1135. B 2202).
2. In the white matter of the nodulus four compartments (XC1-XC4)
are delineated by accumulations of AChE at their borders. Injections in
the caudal de and subnucleus ..13 result in labeled fibers in compartments
XC1 and XC4 that terminated as climbing fibers in the XZr and XZrv zones
fig. 2.1 6 Diagrams of the distribution of climbing fibers in the unfolded flocculus and ventral paraflocculus and diagrams of the inferior olive showing injection sites in cases C 1134 {A) With an injection including cdc. rdc. and vlo. and sparing rostral MAO (B) and case B 2202 (C) with an injection in rostral MAO (D). Note that in C 1134 almost all parts of the flocculus are labeled except for a caudolateral region corresponding to the C2 zone. which receives climbing fibers in case B 2202.
79
[fig. 2.6 !). Zone XZIV is bisected by a central, empty strip of molecular
layer and is wider on the ventral than on the dorsal side of lobule X.
Injections in the rostral de and vlo label zone XZn on the ventral and
dorsal surface of lobule X and a central strip indicated as XZ1v•. A
rostrolateral area in the XZI contains labeled climbing fibers after
injections of subnucleus 13. the caudal de. and the rostral de and the vlo.
This area is indicated with a question mark in figs. 2.6 i. Labeling in XZm
and xzv is in accordance With the data from the retrograde labeling
experiments of Katayama and Nisimaru [1988). XC3 and XZm are labeled
when the injection included the dmcc [K 397, B 2202, C 1135). Labeling
is present in a lateral strip of climbing fibers covering the lateral rim of
the nodulus (XZvl when the injection of the rostral MAO included its
lateral border region [C 1135). but not when this region was spared [B
2202, C1134). Compartment XC3 is wedged between compartments XC3
and XC4 and exists only in the dorsal cortex of the lobule. It is continuous
with a distinct, much wider compartment in the lateral white matter of
lobule lX.
2.5 Discussion
2.5.1 The correlation of white matter compartments with climbing fiber
zones in the floccutus.
The present study showed that fibers from certain subnuclei of the
inferior olive segregate into bundles that occupy specific compartments
within the white matter of the flocculus and the nodulus. They travel
within these compartments until they penetrate the granular layer to
terminate in the molecular layer as climbing fibers on their target
Purkinje cells. Climbing fibers projecting through these compartments
constitute zones that are continuous across the folia of the flocculus and
the nodulus and sometimes extend into adjacent lobules. The graphical
reconstructions of the climbing fiber zones of the flocculus {figs. 2.2; 2.9;
2.16) exemplify the approximately orthogonal relation of the zones with
so
IX ~
Q,,.,
b.,,ll
11ZIJI
'<''!"· . I /~
'J .• ,. __
fip\
IXJ
fig. 2.17 Drawings of labeled fibers in transverse sections through brain stem and cerebellum in case B 2202. Labeling is present in the C2 compartment and zone and in a region corresponding to XC3 in lobules IX and X.
81
fig. 2.18 Darkileld photomicrographs of transverse sections through the flocculus in case C 1134. Note the relatively sparse number of labeled fibers at the borders of the white matter compartments FC~o FCz. and FC3 and the dense labeling in the lateral part of compartment FCz. Labeled fibers are absent from compartment C2 (B). No labeling is present in the position of the nucleo-olivary tract {asterisk in A and B, compare fig. 2.14 B,D).
82
the transverse interfolial fissures. The AChE-positive raphes. which
subdivide the floccular white matter into compartments (chapter l) shift
medially in the direction of the caudal pole of the flocculus. A similar shift
was observed in the position of the climbing fiber zones FZ1-FZrv.
The caudal de, the rostral de, and the vlo, first described by Kooy
(1916). are segments of a continuous cell column located dorsomedial to
the medial accessory olive. The borders between these segments were
never precisely defined With anatomical methods, but they have been
identified in the rabbit by recording field potentials in de and vlo after
flash stimulation of the ipsilateral and the contralateral eyes (Maekawa
and Takeda. 1977; Takeda and Maekawa. 1980) and on the basis of the
laterality and orientation of the best response axis for rotation of the
Visual surround (Simpson et a!.. 1981: Leonard et al.. 1988). The border
between the rostral and caudal de was found to be located at the
constriction of the cell column, which marked the transition between
units in the caudal de activated from the contralateral eye by temporal to
nasal rotation around the vertical axis (vertical axis neurons) and the
more rostral region comprising rostral de and vlo. containing units
activated by rotation around a horizontal axis. This horizontal axis enters
the ipsilateral eye at an angle of about 45° With the midsagittal plane and
leaves the contralateral eye at an angle of about 135°. This axis is
approximately collinear with the axis perpendicular to the contralateral
anterior semicircular canal. Stimulation of the ipsilateral eye dominates
the response of most cells in the rostral de (anterior (45°) axis neurons).
Cells in the vlo and in the most rostral part of the rostral de prefer
stimulation of the contralateral eye (posterior (135°) axis neurons). The
border between the caudal de and the rostral de is in accordance with the
general pattern of afferent connections of the caudal de from the NOT.
the DTN and !TN. and of the rostral de and the vlo from the MTN and the
VTRZ. The electrophystological and morphological criteria of Leonard et
al. (1988) for the subdivision of the de and vlo were applied in our studies
to place our injections of the inferior olive.
Several preVious anatomical studies found a zonal arrangement in
the climbing fiber projection to the flocculus (Yamamoto. l979a. Gerrits
83
and Voogd, 1982, Sato et al, 1983a; Ruigrok et al., 1992). Injections of
HRP in the rabbit flocculus resulted in different patterns of retrograde
labeling In the olive [Yamamoto, 1979a). A few small injections in the
rostral and middle portions of the flocculus resulted in retrograde
labeling mainly restricted to the rostral de. Most injections that -covered
folia fl-f4 showed labeled cells in the rostral de with the vlo and in the
caudal de, separated by an empty segment of the de. This empty segment
of the de and the principal olive contained labeled cells when the
injection included folium p.
Similar observations were made in the combined retrograde
WGA-HRP and anterograde Phasaeolus vulgaris lectin tracing study of the
olivocerebellar projection in the rat (Ruigrok et al., 1992). The caudal de
was found to project to paired FE/FE' zones in the flocculus. A more
rostral segment of the de, corresponding to the "empty segment" of
Yamamoto (l979a), projected to the extension of the FE zone in the
ventral paraflocculus. The vlo innervated two other zones (FD and FD'),
which interdigitate with, and are situated, respectively, caudal to FE and
FE' in the flocculus. More rostrally located cells in the PO projected to an
extension of the FD and FD' !n the ventral paraflocculus. FD and FD' fused
around the end of the FE zone and continued as the D zone in the ventral
paraflocculus. Both Yamamoto (l979a) and Ruigrok et al. (1992)
recognized the projection of the rostral MAO to the caudal flocculus,
which continues in the ventral paraflocculus as the C2 zone. Yamamoto's
(1979a) observations, therefore, indicate that cells projecting to more
rostral segments of the zones FZI-FZrv are located more rostrally in the
caudal de and in the rostral dc/vlo/PO column. Such a shift was not
observed with the relatively large injections in the dc/vlo In our
experiments.
A more intricate pattern in the climbing fiber projection to the
flocculus has been shown by Gerrits and Voogd (1982) !n the cat. Using
an anterograde autoradiographlc technique they found that the caudal de,
rostral de, and vlo each supplied two zones of Purkinje cells in the
flocculus. Their medial Fl zone received a projection from the caudal de
and did not extend into the paraflocculus. Its lateral side was bordered by
84
fig. 2.1 9 Summary diagram of the projection of the de. vlo and rostral MAO through FC1-FC4 and Cz to the corresponding zones (FZ) in the flocculus.
the F2 zone, which received fibers from the vlo. The Fl/F2 border was
considered as genuine because it separates climbing fibers from
non-contiguous portions of the inferior olive. The Fl zone of the cat,
therefore, is the equivalent of the FZrv zone in the rabbit and the FE' zone
in the rat. The via-innervated F2 zone in the cat is bordered on its lateral
side by F3. which receives a projection from the rostral de. The same
pattern is repeated more laterally. with the appearance of F4 (caudal de).
F5 (vlo) and F6 (rostral de). The vlo and rostral de zones continued
across the posterolateral fissure into the adjoining medial extension of
the ventral paraflocculus (ME), where they enclose the end of the caudal
de-innervated F4 zone. This pattern is very similar to the situation in the
rabbit and the rat. and. consequently. F4 of the cat can be considered as
the equivalent of our FZn zone and the combined F2/3 and FS/6 zones as
the homologues of the FZm and FZr zones of the rabbit flocculus. A
subdivision of the FZ1 and FZm zones was not observed in our material,
although an additional AChE-positive raphe sometimes was present,
which divided the FC1 compartment over some of its extent (Chapter 1).
The studies of Yamamoto (1979a). Gerrits and Voogd (1982, 1989),
Ruigrok et al., (1992) and our own observations lead to the conclusion
85
that an identical pattern of climbing fiber connections is present Ln the
flocculus of different mammalian species and that this pattern finds its
expression in the compartmental subdivision of the white matter of the
flocculus of the rabbit as revealed in AChE-processed brain sections
(Chapter 1).
Sato et al. (1983) distinguished three zones in the cat flocculus,
with projections from the rostral de and the vlo to rostral and caudal
zones and projections from the caudal de to a middle zone. These zones
resemble our FZI. FZu, and FZm. They did not recognize the projection
from the caudal MAO to the C2 zone of the flocculus nor a second zone
receiving from the caudal de. A similar trizonal organization was also
found in the flocculus of the monkey (Balaban et al.. 1981)
2.5.2 The correlation of white matter compartments and the climbing Jiber
projection to the nodulus.
Our obsenrations on the projection of the inferior olive to the
nodulus confirm the description of Katayama and Nisimaru (1988), based
on retrograde axonal transport of HRP and resemble the pattern reported
by Balaban and Henry (1988). also in the rabbit. We distinguished four
compartments in AChE-stained material in each half of the white matter
of the nodulus. A paramedian, l mm wide XZ1 zone was innervated
through compartment XC 1 by the subnucleus J3 and the caudal de. The
contributions of these two subnuclei could not be distinguished in our
experiments. Katayama and Nisimaru (1988) subdivided this zone into a
medial .6-innervated zone and a lateral caudal de-innervated zone. We
distinguished a triangular rostrolateral area bordering on XZr on the
ventral surface of the nodulus, which received projections from the caudal
de and/or the nucleus J3 and the rostral dc/vlo. It was indicated with a
question mark in figs. 2.6 i. Olivocerebellar fibers from the rostral dc/vlo
occupy compartment XC2 and innervate the XZn zone, which was also
recognized by Balaban and Henry (1988) and Katayama and Nisimaru
(1988). Compartment XC3 and the XZm zone are only present on the
86
dorsal surface of lobule X and are innervated by the dmcc. XZIV receives
projections from the caudal de and, in a central strip, XZyv•. from the
rostral dc/vlo. Our diagram differs from that of Katayama and Nisimaru
(1988) because, on the dorsal surface of the nodulus, XZw
(corresponding to their zone iv) intervenes between XZm and xzv (their
zones v and vi), which are innervated by the dmcc and the rostral MAO,
respectively.
Climbing fibers to the flocculus and nodulus originate from different
regions of the rostral MAO. Cells projecting to the flocculus of the rabbit
are located in the rostral tip of the MAO {Alley et al.. 1975). Projections
to the nodulus of the rabbit are derived from cells located in the lateral
margin of the rostral MAO (see fig. 2 D of Katayama and Nisimaru, 1988)
The differential origin of projections to the C2 zone in the flocculus and
the XZm and XZv zones of the nodulus is supported by our experiments.
Injections that spared the lateral half of the rostral MAO, but inducted the
dmcc {K 397. B 2202) caused labeling in XZm but not in XZv. Injections
that included the rostral tip of the MAO {C !!35. B 2202) labeled
climbtng fibers of the C2 zone.
Descriptions of the zonal arrangement of the olivocerebellar
projection to the nodulus in other species are less complete and will not
be discussed. The continuity of the X.Zy, XZn. XZm and XZrv zones with
similar zones in the uvula is obvious from studies of this lobule in different
species (Brodal. 1976; Groenewegen and Voogd, 1977: Groenewegen et
al., 1979: Kawamura and Hashikawa, 1979: Eisenman, 1984; Sato and
Barmack. 1985. rabbit; Bernard. 1987; Kanda et aL. 1989. cat: Apps.
1990. rat).
The anatomical and electrophysiological studies of Takeda and
Maekawa {1984a.b; 1989a.b), Maekawa et al.. (1989), Kusunoki et aL.
(1990) and Kano et al. {1990) in the rabbit showed branching of climbing
fibers between flocculus and nodulus. Takeda and Maekawa ( 1989) found
that 35.6% of the climbing fibers In the flocculus and 63.4o/o of the
climbing fibers in the nodulus that could be driven by optokinetic stimuli
could be activated by their collaterals, terminating in the complementary
87
lobule. Retrograde labeling from flocculus and nodulus with different
fluorescent tracers resulted in double-labeling of 9-27% of the de neurons
and 12-18% of the cells of the vlo (Maekawa et a!., 1989). There was no
distinct spatial segregation of the cells projecting to the nodulus and the
flocculus. Branching of olivocerebellar fibers could not be observed in our
experiments, but a continuity was observed between the white matter of
the flocculus and the nodulus. Fibers from the rostral dc/vlo. contained
in FC 1 , and from the caudal de, contained in FC2. continue from the
flocculus, along the attachment of the roof of the fourth ventricle {t4 in
fig. 2.3 k). into the nodulus. lt seems likely that climbing fibers
innervating the paired zones FZI/Fm and FZu/FZIV of the flocculus are
branches from the same axon, but this has not been definitely proven.
2.5.3 The laterar A-zone and the nucleo-olivary pathway in the rabbit.
Additional observations in our study concern the presence of a
lateral A zone in the medial hemisphere of the posterior lobe {simple,
ansiform and paramedian lobules). innervated by the lateral part (group c)
of the caudal MAO. The olive-cerebellar fibers in the white matter collect
around and probably innervate the dorsolateral protuberance of the
fastigial nucleus (K 369 and K 314, figs. 2.3: 2.5) before they enter the
white matter of the lobules. The lateral A zone and its climbing fiber
projections from the group c of the caudal MAO were first distinguished
by Buisseret-Delmas (1988) and Akaike (1986a.b. 1987, 1992) in the rat.
The projection of the lateral A zone to the dorsolateral protuberance was
substantiated by Buisseret-Delmas (1988).
The course of the nucleo-olivary fibers, first described by Legendre
and Courville (1987) in the cat, could be confirmed because this tract was
retrogradely labeled from WGA-HRP injections of the rostral de, the vlo,
and portions of the rostral MAO. DAO and the PO (but not from the caudal
de and the caudal MAO) in our experiments. The nucleo-olivary fibers
could not be labeled with injections of tritiated leucine in the inferior
olive. The nucleo-olivary fibers leave the nuclei and pass through and
around the parvicellular part of the lateral cerebellar nucleus in the
88
rostral wall of the lateral recess. They ascend in the lateral angle of the
fourth ventricle dorsal to and intermingling with the fibers of LOwy's
bundle to become located ventromedial to the superior cerebellar
peduncle. Subsequently they enter the decussation of this tract.
2.5.4 Functional considerations.
The differential projection of the dc/vlo complex to the different FZ
in the flocculus is part of the neuronal circuitry that is responsible for the
control of eye movements. With the use of microstimulatton three
functional zones (indicated as "microzones" by Ito. 1984) were
distinguished in the rabbit flocculus either by recording of eye
movements (Dufosse et a!., 1977; Nagao et a!., 1985, chapter 4) or by
recording the effects of vestibula-ocular reflexes on evoked eye muscle
activity (Yamamoto, 1979b; Ito et a!., 1982b). Stimulation of a narrow
strip in the rostral flocculus resulted in horizontal eye movements or
suppression of VOR's from the horiZontal canal (Horizontal (H) zone li).
Stimulation rostral and caudal to zone II inhibited VOR's from the
anterior canal or provoked vertical eye movements (rostral and caudal,
Vertical M zones I and III). Stimulation points of the H and V zones
showed considerable overlap. Moreover, inhibition of the VOR for the
contralateral inferior oblique muscle (Ito et al., 1982b) and rotatory eye
movements could be produced by stimulation from an area in the ventral
flocculus that ran obliquely through zones 1-111 (Rotatory (R) zone, Nagao
et al., 1985). Zone II corresponds to the more caudal of two strips of
Purkinje cells in the rabbit flocculus that project to the medial vestibular
nucleus (Yamamoto and Sh!moyama (1977). Zones I and Ill correspond
to the Purkinje cell strips projecting to the superior vestibular nucleus
(Note that the rostralmost zone of Purkinje cells. projecting to the medial
vestibular nucleus and described earlier by these authors was no longer
mentioned In their later work [Ito, 1984)).
A zonal pattern has also been indicated by local electrical
stimulation in the rabbit flocculus. Four zones have been identified in this
way (Nagao et al., 1985): one dorsoventrally elongated zone from which
89
horizontal eye movements were evoked (H zone) and two flanking zones
from which vertical eye movements were elicited (V zones). An R zone
from which rotatory eye movements were evoked ran almost
perpendicularly through the H and V zones. It was asserted that the zones
of Purkinje cells controlling relay cells of different VORs matched the eye
movement zones. In chapter 4 the relationship between the white matter
compartmentation, climbing fiber zones and eye movements will be
demonstrated in detail. Also in this study three zones were associated
with eye movements. but these ran parallel and did not intertwine.
Climbing fiber responses (complex spikes} to optokinetic
stimulation were recorded from Purkinje cells of the flocculus (Kusunoki
et al., 1990) and the nodulus [Kano et al., 1990) of the rabbit. Cells that
preferred horizontal movement were activated by a temporally to nasally
moVing optokinetic stimulus presented to the ipsilateral eye [!psi F cells).
These cells were located in a single, 1 mm wide strip along the caudal
part of the rostral one third of the flocculus. Some Ipsi F cells were found
scattered in folium p. There is a good correspondence between this strip
and the projection of the caudal de, which contains the vertical axis
neurons of Leonard et al. [1988), to the FZn zone. The rostral and medial
FZrv zone, with a similar climbing fiber input, was not noticed by
Kusunoki et al. (1990).
Most of the cells that preferred vertical optokinetic stimulation
were excited by upward movement presented to the ipsilateral eye {Ipsi
U cells) or downward movement presented to the contralateral eye
(Contra D cells). Cells with a preference for vertical movement were
located in two strips, one rostral and one caudal to the !psi F cells. These
strips correspond to the FZy and FZni zones, which receive climbing fiber
projections from the rostral de and the vlo. The optokinetic stimulus
activating the !psi U cells appears to be equivalent to the rotation that
activated the posterior (135°) units in the vlo in the experiments of
Leonard et al. (1988). The Contra D stimulus probably was relayed by the
anterior (45°) units of Leonard et al. (1988) located in the rostral de.
There is a tendency for Ipsi U cells to be located more rostrally and
medially in the flocculus than the Contra D cells (Kusunoki et al., 1990,
90
their Fig. 10 B. C). Similarly, the projection of the vlo to the F2 and F5
zones is medial With respect to the projection of the rostral de to F3 and
F6, as noticed by Gerrits and Voogd (1982) for the flocculus of the cat.
Such a gradient could not be observed in the corresponding zones in the
rat {Ruigrok eta!., 1992) or the rabbit.
The Purkinje cell zones in the rabbit flocculus, defined by
retrograde labeling (Yamamoto and Shimoyama, 1977), microstimulation
(Dufosse et a!., 1977; Yamamoto. 1979b; Ito et al.. 1982a.b; Nagao eta!.,
1985} or optokinetic stimulus-induced complex spikes activity {Kusunoki
et al.. 1990) overlap considerably, especially in the ventral folia of the
flocculus. In this respect they differ from the discrete climbing fiber
projections identified in this study on the basis of the compartmental
subdivision of the floccular white matter.
Purkinje cells responding with complex spikes to optokinetic
stimulation have also been identified in the nodulus of the rabbit
cerebellum (Simpson and Alley, 1974; Kano et al., 1990). !psi F cells
were found in medial and lateral strips on the ventral surface of the
nodulus. These two strips flanked a strip of Ipsi U and Contra D cells.
which extended to the dorsal surface of the lobule (Kano et a!.. 1990).
This arrangement approximately corresponds to the pattern of two caudal
de (and/or subnucleus J3-) innervated zones, flanking a single zone of
climbing fibers from the rostral dc/vlo, as found by Katayama and
N!simaru (!988: zones!!, 1!1 and iv), Henry and Balaban (!988) and by us
(lateral part of xz,. XZn and XZIV• zones). A few Purkinje cells of the !psi
D or Contra U type were found in the medial, subnucleus j3-innervated
part of zone XZ1 of the nodulus and the uvula (Kano eta!., 1990).
Our data indicate that the pattern of terminal zonation of the
climbing fiber projection to the flocculus is rooted in the anatomical
compartmentation found with AChE histochemistry. This compartmen
tation provides an intrinsic, independent framework that is a useful tool
for correlating anatomical and physiological data. This compartmentation
is apparently the basis of the modular organization of differential control
by the flocculus of eye muscle act!vlty.
91
92
Chapter 3
Zonal organization of the flocculo-uestibulor nucleus projection in
the rabbit. R combined WGR-HRP tracing and acetylcholinesterase
histochemical study.
3.1 Introduction
The flocculus plays an important role in the control of eye
movements by processing Visual and vestibular information (Robinson.
1976; Ito et al., 1982a: Nagao 1983: Ito. 1984). Via their projections to
the vestibular nuclei Purkinje cells of the flocculus inhibit certain groups
of vestibulo-ocular relay cells (Ito eta!., 1970: Fukuda et al., 1972; Baker
et al.,1972; Highstein, 1973; Kawaguchi, 1985; Sato and Kawasaki, 1987.
1990a,b, 1991; Sato et a! .. 1988). The projection of the flocculus to the
vestibular nuclei has been known since the studies of Dow {1936,
1938a,b, 1939) with the Marchi method in rat. cat and monkey. Recent
studies using modern tracing techniques showed that the flocculus sends
major projections to the superior {SV) and medial vestibular nucleus
(MV), group y, lateral cerebellar nucleus (LCN), basal interstitial nucleus
and minor projections to the descending vestibular nucleus (DV) and the
nucleus prepositus hypoglossi (Jansen and Broda!, 1940: Voogd, 1964:
Angaut and Broda!, 1967: Alley. 1977; Haines, 1977; Yamamoto and
Shimoyama. 1977: Balaban et al., 1981: Sato et al., !982a and b:
Carpenter and Cowie, 1985: Langer et al., 1985b; Epema, !990). Some of
these studies have indicated that the Purkinje cell population of the
flocculus consists of subsets. which are arranged in zones that cross the
folia of the flocculus and project to different vestibular nuclei.
Yamamoto and Shimoyama (1977) found Purkinje cells of the rabbit
flocculus that project to the MV and SV to be distributed in a striped
pattern. They distinguished two Purkinje cell zones projecting to MV and
two zones to SV. The two zones projecting to MV interdigitate With and
are located rostral to the zones projecting to SV. Purkinje cells of a fifth
93
zone, located in the caudal flocculus and also extending in folium p. were
labeled from injections of HRP in the cerebellar lateral nucleus
(Yamamoto, 1978). In her later publications {Yamamoto, 1978. 1979a,b)
the presence of the rostral MV zone no longer was mentioned. A single
:MV zone, flanked by two SV-projecting zones also was distinguished in
monkeys (Balaban et a!.. 1981). A different pattern of efferent
connections was described for the three zones in the flocculus of the cat
by Sato et al. (l982a and b). Their rostral zone was connected with SV.
their middle zone with MV and the caudal zone projected to group y and
sparsely to the ventromedial, parvicellular lateral cerebellar nucleus.
Similar zonal patterns were revealed by microstimulation of the
rabbit flocculus and recording of the inhibition of specific vestibule-ocular
reflex pathways (Yamamoto, 1979b; Ito et al.. l982b). Local electrical
stimulation elicited eye movements (Dufosse et al., 1977; Balaban and
Watanabe, 1984; Sato et al.. 1984, 1990a. 199!; Nagao et al., 1985;
Belknap and Noda. 1987). In the rabbit microstimulatlon in the Purkinje
cell layer of the flocculus resulted in horizontal. vertical and rotatory eye
movements located in four zones (Dufosse et al.. 1977: Nagao et al..
1985). An H-zone, in which electrical stimulation elicited horizontal eye
movements was flanked by two V-zones in which stimulation elicited
vertical eye movements. This trizonal organization of eye movement
control has also been found in cat (Sato and Kawasaki. 1984. 1987,
1990a,b, 1991: Sato et al., l982a, b. 1983a. 1984, 1988) and monkey
(Balaban and Watanabe, 1984). In the rabbit the zonal organization was
complicated by the presence of rotatory eye movements on stimulation of
an R-zone which overlapped large parts of the H- and V -zones (Ito et al..
l982b. Nagao et al.. 1985).
The functional representation of eye movements in zones probably
reflects the zonal arrangement in the efferent and afferent (climbing
fiber) connections of the Purkinje cells. Precise correlations of the
functional and anatomical properties of the floccular zones are hindered
by inconsistencies between the different anatomical and physiological
studies on the number and extent of the zones, by large indiVidual
variations in the number and size of the folia of the flocculus and by
94
differences in zonal organization between different' species.
Recently, a morphological zonation in the rabbit flocculus has been
demonstrated using acetylcholinesterase (AChE) histochemistry (Tan et
al., 1989, see also chapter 1). AChE-positive raphes subdivide the white
matter of the flocculus into 5 compartments. In the preVious study
(Chapter 2) the AChE-compartmentation was used as an independent
framework to correlate data from experiments on the olivocerebellar
projection to the rabbit flocculus. The climbing fibers terminate in 5
zones, FZI-FZIV and C2, each of which is associated with a particular
white matter compartment (FC1-FC4. C2). The rostral dorsal cap (rdc)
and adjacent ventrolateral outgrowth (vlo) supply climbing fibers to FZ!
and FZm via compartments FC 1 and FC3. the caudal dorsal cap (cdc)
supplies FZn and FZIV via FC2 and FC4. and the rostral medial accessory
olive innervates zone c2 via compartment c2.
In this study we investigated with anterograde and retrograde
transport methods the differential projection of Purkinje cells of the
floccular zones FZI-FZrv to the vestibular and cerebellar nuclei and the
relation of their axons to the white matter compartmentation, as
described in chapter 1 and Tan et al. (1989). Recently. Purkinje cell
axons from identified zones were also traced with biocytin by De Zeeuw et
al. (1992).
95
3.2 Materials and methods
3.2.1 Retrograde studies.
HRP (33%) dissolved in phosphate buffered saline was pressure
injected within the vestibular nuclei of rabbits with a l )lL Hamilton
syringe with a 25 gauge needle (Epema. 1990). Two of these cases v;~ll be
presented: one with an injection of SV (case C 253 (0.1 ;;LII and one in
the MV (C 480 (1.0 ;;LII. After a survival time of two days the animals
were heparinized and transcardially perfused under deep anaesthesia
with 0.5 L saline, followed by 1.5 L citrate buffer (0.1 M, pH 7 .4)
containing 1% formaldehyde and 1.25% glutaraldehyde and subsequently
washed with 0.5 L citrate buffer (0.1 M, pH 7 .4) containing 8% sucrose
{Mesulam, 1978). The brains were removed, embedded in 10% gelatin
(Voogd and Feirabend. 1981), and transversely sectioned at 40 ;;m. A
series containing one out of every four sections was incubated with
diaminobenzidine (DAB; Graham and Karnovsky, 1966) and
counterstained with cresyl violet.
3.2.2 Anterograde studies.
ln 9 pigmented Dutch belted rabbits (K 227. 244, 301. 305, 340,
355, 358, 360, 370) small quantities (5-15 nl) of a 7% solution of
horseradish peroxidase coupled to wheat germ agglutinin (WGA-HRP.
Sigma type VI) were injected with air pressure into the flocculus. Further
processing of the transversely sectioned brains for TMB, DAB and AChE
has already been described in chapter 2.2 and 3.2. In some cases (K 301,
305, 340, 355. 360, 370) the injection sites were determined by
recording climbing fiber responses of the Purkinje cells to rotating
random dot patterns (Graf et al., 1988; Leonard et al.. 1988). Injection
sites are indicated in diagrams of the unfolded flocculus (see fig. 3. 1).
When the flocculus was unfolded, the length of the Purkinje cell layer in a
single transverse section was indicated as a straight line. The sections
were aligned using the distance from the most ventromedial point of the
96
floccular Purkinje cell layer (B in fig. 3.1) to the midline (M). The
distance between the sections (t) correspond to the true distance
between the reconstructed sections multiplied by the magnification factor
(fig. 3.1). The location and extent of the injection sites, the interfolial
fissures and the approximate position of the climbing fiber zones
(FZr-FZrv. C2). estimated by extrapolating the AChE raphes to the
Purkinje cell layer, were indicated in these diagrams (figs. 3.6; 3.8). The
labeled fibers in the cerebellum and brain stem were plotted. In the
white matter of the flocculus the raphes were identified in adjacent AChE
sections. The subdiVision and nomenclature of the vestibular complex
used Ln the present study is according to Epema et al. (1988).
ml! C2 ZOM u~d compor~m~n~
~ FZJ and FC,
~ FZJL ond Fez
FZm and FC3
[:J . FZ:u ond Fe, L .. --//- -
fig. 3. 1 Diagram of the method used to unfold the cortex in preparing maps of the unfolded surface of the flocculus showing the approximate extent of the floccular zones. The Purkinje cell is straightened; the position of the injection site and the approximate position of the cortical zones FZI- FZIV and C2 is indicated. Sections are aligned using the distance between the midline (M) and the most ventromedial point of the Purkinje cell layer (B). Four sections (A-B to A'"-B"') are reconstructed in C. The distance between the reconstructed sections (t) is obtained by multiplying the distance between the reconstructed sections by the magnification factor. Symbols indicating the floccular zones and compartments are shown on the right.
97
M
3.3 Results
3.3.1 Retrograde experiments.
in C 480 the HRP injection involved all parts of the MV and the
ventromedial portion of the LV. Retrogradely labeled Purkinje cells are
found in all folia of the flocculus (flg. 3.2). Two distinct zones of Purkinje
cells were labeled. The one located rostrally and medially contained only
a few labeled cells. The second zone was located more laterally and
caudally and remained separated from the rostromedial zone by a band of
unlabeled cells. lt extended into the rostral tip of folium p (Fig. 32 d-e).
More caudally, scattered groups of labeled Purkinje cells were present in
folium p. Mossy fibers and climbing fibers in the flocculus remained
unlabeled. Thick. retrogradely labeled Purkinje cell axons could be traced
from the labeled Purkinje cells, through the granular layer into the white
matter. Each Purkinje cell zone was associated with labeling in a certain
part of the white matter. Labeled axons associated with the medial zone of
labeled Purkinje cells were present along the rostromedial border of the
flocculus With the brachium pontis (figs. 3.2 b-e; 3.4). More caudally,
they assembled in a bundle at the ventromedial tip of the flocculus. dorsal
to the cochlear nuclei (fig. 3.2 d-e). Fibers associated With the lateral
zone of labeled Purkinje cells extended from folium p dorsally into folium
f2 ventrally (fig. 3.2 d). More caudally they collected into a triangular
bundle, With its base resting on the cortex of fl (figs. 3.2 e: 3.4 B). The
topography of the labeled compartments corresponds With the FC4 and
FC2 compartments delineated in AChE-incubated tissue sections (chapter
1.3). In the caudal flocculus the labeled fibers no longer could be
distinguished as two separate compartments. They entered the floccular
fig. 3.2 DraWings of retrogradely labeled Purkinje cells and their axons in transverse sections through the flocculus and floccular peduncle of case C 480 With an HRP injection site involving the rostral part of the medial vestibular nucleus. Most fibers occupy compartment FC2 and FC4, with labeled cells in the corresponding zones FZu and FZrv.
98
a
///~,
j~sc-r bp \ \ r-
,IV ~ j g 480·43
h 480-40
d 480·50
e 480-47
J 480-34
99
a 253-20
b 253-17
f 253-8
100
peduncle in bundles, passing between unlabeled Purkinje cell axons of other compartments. The majority of the retrogradely labeled fibers could
be traced, through the floccular peduncle and group y, to the injection
site in the MV nucleus. Labeled fibers in the ventral part of the floccular
peduncle. immediately dorsal to the restiform body, were tightly packed while the rest of the fibers in the floccular peduncle were more loosely
arranged. Some of these fibers entered the dorsal part of the cochlear
nucleus. Other bundles of labeled fibers arched through the ventral part of
the lateral cerebellar nucleus {fig. 3.4 C). Most of these fibers passed
dorsal to the parvicellular part of the lateral nucleus. More medially they
turned sharply ventrally to reach the MV through the caudal pole of the
SV. A great number of retrogradely filled axons from labeled Purkinje
cells in the anterior and posterior vermis passed through the fastlgial
nucleus and the medial portion of the anterior and posterior interposed
nuclei in the direction of the injection site {fig. 3.2 f-j).
In case C 253 the injection site was situated in SV and LV, touching
upon the MV {fig. 3.3 d-1). The injection extended into the ventral part of
the superior cerebellar peduncle. Retrogradely labeled Purkinje cells in
the flocculus were distributed in two wide zones. separated by an empty strip {fig. 3.5). Purkinje cells located along the rostromedial edge of the
flocculus remained unlabeled. Labeling of Purkinje cells of both zones
extended into folium p {fig. 3.3 c-fl. Retrogradely filled Purkinje cell
axons traveled in two bundles in the rostral white matter of the flocculus
(fig. 3.3 c). The medial bundle remained separated from the medial
border of the flocculus by an empty space, corresponding to FC4 which
contallned labeled axons in the preVious case with an injection of MV.
Dorsally in the white matter of folium p the two bundles became fused,
fig, 3.3 Drawings of retrogradely labeled Purkinje cells and their axons in transverse sections through the flocculus and floccular peduncle of C 253 With an HRP injection site involVing the superior vestibular nucleus. Most Purkinje cell axons occupy compartments FC 1 and FCs with labeled neurons in zones FZI and FZnJ. An extension of the injection site, involVing the superior cerebellar peduncle {d) causes retrograde labeling of many neurons in the cerebellar nuclei (e-i).
!01
c 102
ventrally they were separated by a triangular area containing a few labeled fibers and corresponding to the FC2 compartment (fig. 3.3 d-g; 3.5 C).
The most lateral parts of the white matter of the folia fl-12 remained free
of labeled axons (fig. 3.3 e-g). This region corresponds to the C2
compartment. From the caudal flocculus labeled fibers proceeded in the
floccular peduncle in bundles, passing between unlabeled fibers of the
floccular white matter. They passed through dorsal group y to reach the
injection site in SV (fig. 3.5 C). Other fibers followed a dorsal route,
through the lateral cerebellar nucleus. Their final course was difficult to
establish, because thick, retrogradely filled axons of the superior
cerebellar peduncle and retrogradely labeled cells dominate the picture
of the lateral and interposed nuclei. Purkinje cells of the vermis and their
axons were also retrogradely labeled from this injection site.
3.3.2 Anterograde experiments.
The cases were grouped together according to the white matter
compartment in which the labeled fibers were located and according to
the extent and location of the injection sites in the flocculus compared With the predicted position of the FZ bands.
Group l (K 244, K 355, K 358, K 360)
In this group small injections of WGA-HRP in the cortex of the
flocculus resulted in labeled fibers which clearly grouped together in
compartment FC1 (figs. 3.7 A-B; 3.9; 3.!1). The injection sites inK 244
and K 355 are situated Within the predicted borders of FZ1 (fig. 3.8 B-C),
the injections inK 358 and K 360 encroach upon FZn (figs. 3.6 C; 3.8 D).
The injection site in K 244 is located in f3-f4 and extends more
caudally than in the other three cases (figs. 3.8 B; 3.9 c). The labeled
fig. 3.4 Brtghtfield photomicrographs of transverse sections through the flocculus in C 480. Compare fig. 3.2, d, e and g.
103
c
104
fibers are located In the lateral half of compartment FC,. No labeled fibers
were present in any of the other compartments. Labeled neurons in the
inferior olive are almost restricted to the vlo and the rdc (fig. 3.16). In
the caudal flocculus fibers in this compartment pass medially, dorsal to
compartment FC2. where they subdivide into two fiber streams. One
stream curves dorsalwards through the lateral cerebellar nucleus, turning
medially and caudally along the dorsal border and through the
parvicellular lateral cerebellar nucleus. Within the anterior interposed
nucleus they turn abruptly ventrally to enter the caudal pole of the
superior vestibular nucleus (SV) (fig. 3.9 e-f). A few fibers continue along
the lateral vestibular nucleus (LV) and the descending vestibular nucleus
(DV), close to the restiform body. The second stream remains in a ventral
position and enters the floccular peduncle. It comprises the bulk of fibers
of compartment FC l· Labeled fibers ramify among the cells of group y
located between the white matter of the flocculus, restiform body and the
cochlear nucleus (fig. 3.9 f; 3.10 B. C) and between other cells of the
dorsal group y, situated between the fibers of the floccular peduncle. Due
to the great number of labeled fibers traversing through dorsal group y it
cannot be established with certainty, whether the axons actually
terminate on these neurons. Labeled fibers are scarce in ventral and
dorsal parts of the floccular peduncle. A few labeled fibers pass through
the ventral group y. along the dorsal border of the restiform body. Very
few fibers passing through the caudal pole of SV enter LOwy's bundle.
Labeled fibers of the second stream intersect With fibers arching through
the lateral and interposed nuclei In the region of the caudal pole of SV
{fig. 3.9 e). Most labeled flbers terminate mainly in the dorsolateral and
central parts of SV, and in its caudal pole (fig. 3.9 b-e). A few fibers can
be traced into MV and DV. Retrogradely labeled cells of different size are
bilaterally present in SV, MV and PH. Some retrogradely labeled cells are
present caudally in the ipsilateral dorsal group y and in the region
fig. 3.5 Brtghtfleld photomicrographs of transverse sections through the flocculus in C 253. Compare fig. 3.3. b, e and f.
!05
A
-=11111'1
0 - =lnHn
fig. 3.6 Diagrams of the unfolded flocculus (left) and a transverse section through the flocculus at the level of the injection site (right) in cases K 301, K 305, K 360, K 370. The extent of the injection sites is shown in relation to the floccular compartments and zones. For explanation of the symbols indicating zones and compartments. see fig. 3.1.
106
fig. 3. 7 Darkfield and brightfield photomicrographs of TMB processed (A.C.El and adjoining AChE processed sections through the flocculus showing the position of labeled fibers in the different white matter compartments. K 360 with an injection site in FZr and anterogradely labeled fibers in FC1. K 301 with an injection site in FZn and labeled fibers in FCz, K 370 with an injection site in FZm and labeled fibers in FCs.
caudall and dorsal to the parvicellular lateral cerebellar nucleus.
Retrogradely labeled cells extend in the roof of the fourth ventricle to the
lateral border of the nodulus.
The injection site in K 355 is located in f3 in the center of the
estimated FZ1 zone (fig. 3.8 C). The majority of the retrogradely labeled
cells in the inferior olive is located in the vlo and rdc {fig. 3.16). Labeled
107
0 -.,lmm
ventral
_ ~lmm
108
fibers in the cerebellar white matter are mainly located in the medial half
of compartment FC 1 With a few fibers in FC2. The course of the labeled
fibers is the same as in case K 244.
The injection sites in K 360 (fig. 3.6 C) is situated in FZ!. but
encroaches upon FZn. The majority of the retrogradely labeled cells in
the olive is located in vlo and rdc, but a substantial number is found in the
cdc. Labeled fibers are located in FC! and central FC2 (figs. 3.7 A, B: 3.11
c-e). Tv.ro fiber streams, one arching through the lateral and interposed
nuclei and one passing through the floccular peduncle can be
distinguished. Most fibers terminate in dorsolateral, central, and caudal
parts of SV (fig. 3.11 d-f). Other fibers, passing between LV and the
restiform body, Uiwy's bundle and dorsal DV distribute to MV (fig. 3.11
g-j). They terminate in the parvicellular part of MV, dorsolateral to the
acoustic stria, among larger cells of MV, ventrolateral to the stria, and in
more caudal parts of this nucleus. Very few labeled fibers can be seen in
the nucleus prepositus hypoglossi. Diffuse labeling is present in the
medial part of the dorsal cochlear nucleus (fig. 3.11 h-i). Retrogradely
labeled neurons are present bilaterally in the vestibular nuclei except for
LV. The ipsilateral dorsal group y contains a substantial number of
retrogradely labeled neurons, very few are present contralaterally (fig.
3. II f-g). Similar observations were made in K 358 (fig. 3.8 D)
Group 2 (K 301, K 305, K 340)
The injection sites of group 2 are located in FZu and labeled
Purkinje cell axons travel in compartment FC2. In case K 301 and K 340
retrograde labeling in the contralateral inferior olive prevails in the cdc
(fig. 3.16). The caudal pole of the inferior olive in K 305 was not
fig. 3.8 Diagrams of the unfolded flocculus (left) and a transverse section containing the injection site (rtght) in cases K 227, K 244, K 355, K 358. K 340. The extent of the injection sites are shown in relation to the floccular compartments and zones. For explanation of symbols indicating zones and compartments see fig. 3.1.
109
f 244·81
fig" 3.9 Drawings of labeled fibers and neurons in transverse sections through the flocculus and brain stem in case K 244 with an injection in FZr. Note the location of efferent fibers in FC1. Fibers follow two routes. one through the floccular peduncle and group y. the other passing more dorsally through the lateral and anterior interposed cerebellar nuclei. Fibers terminate in mainly the superior vestibular nucleus (SV).
1!0
sectioned, but no labeled cells were present in the vlo in this case.
The injection site in K 301 in the rostral pole of folium p [figs 3.6
A; 3.12 d) labels a small number of fibers in dorsal FCz (fig. 3. 7 C. D).
They pass dorsal to the parvicellular part of the lateral nucleus or ventral
to it, through the floccular peduncle (fig. 3.12 e-fi. In the vestibular
nuclei they terminate in MV, medial and lateral to the acoustic stria (fig.
3.12 f-h}. A few labeled cells are present bilaterally in dorsal and ventral
group y (fig. 3.12 f).
The injection site in K 305 is located in the FZn zone of fz (fig. 3.6
B. 3.13 C). The labeled fibers occupy most of the compartment Fez
except for a dorsal triangle (fig. 3.13 c-e). This triangle becomes filled
after an injection in FZu in folium p (K 301. figs. 3. 7 C: 3.12 d). A few
labeled fibers are present in FC3. At the level of the floccular peduncle
these fibers follow a similar course as fibers in the previous experiments.
Two fiber streams can be observed. A small number of fibers curves
through the lateral cerebellar and anterior interposed nuclei and enters
the vestibular nuclei through the caudal SV, lateral to LV. The majority of
the fibers accumulate in the floccular peduncle and travel medially in
Uiwy's bundle to the lateral angle of the fourth ventricle where they bend
sharply in a ventral direction to enter I\W. Labeled fibers in the floccular
peduncle traverse the dorsal and ventral divisions of group y, but no
distinct signs of termination could be observed. The majority of the
labeled fibers terminates in MV. They are present in the rostral MV,
ventral to SV, and more caudally they are especially numerous
dorsomedial and ventrolateral to the acoustic stria (fig. 3.13 c-i). Labeled
fibers extend beyond the stria into the caudal MV. Some fibers enter SV.
Diffuse labeling is present in the medial part of the dorsal cochlear
nucleus (fig. 3.13 h). A few labeled fibers were observed in the nucleus
prepositus hypoglossi. Some retrogradely labeled neurons are present
bilaterally in all vestibular nuclei, with the exception of LV. Ventral and
dorsal group y contains some labeled neurons, a few labeled neurons are
present in contralateral group y. Identical observations were made in case
K 340 (fig. 3.8 E).
111
c - • !12
Group 3 (K 227, K 370).
The injection sites in both cases are located in FZm (figs. 3.6 D; 3.8
A). In K 227 the injection extends into FZn and also involvement of FZrv
cannot be excluded. Labeling in the olive prevails in the vlo and the rdc in K 370. In K 227 retrogradely labeled cells ares almost equally distributed
over the cdc and tbe rdc wJtb tbe vlo (fig. 3.16).
Labeled fibers inK 370 are almost restricted to FC3 (fig. 3.7 E, FJ.
In the caudal flocculus these fibers enter the floccular peduncle as a
single compact bundle. They ramifY among cells of dorsal group y. Labeled
fibers are also present in ventral and caudal parts of the floccular
peduncle, a few fibers travel through ventral group y, bordering on the
restiform body (fig. 3.14 e-g: 3.15 C-D). Most fibers enter and terminate
in tbe dorsolateral, central, and caudal parts of SV (fig. 3.15 A-B). Few
fibers travel tbrough L6wy's bundle and through DV towards MV, where
they terminate in the region surrounding the acoustic stria, and in more
caudal parts of MV. A considerable number of labeled fibers appears to
terminate in the dorsal cochlear nucleus (fig. 3.14 h-i). A few labeled
fibers were seen in the nucleus prepositus hypoglossi. Retrogradely
labeled cells are present bilaterally in SV, DV and MV. The dorsal and
ventral group y contain a few retrogradely labeled cells; more labeling is
observed around and caudal to the parvicellular lateral nucleus. A few
labeled cells are present In the contralateral group y. Small labeled
neurons in the roof of the fourth ventricle extend, ventral to the
cerebellar nuclei, into the white matter of tbe nodulus (fig. 3.14 g-h).
Labeling in case K 227 is a combination of the distributions in
groups 2 and 3. Fibers terminate in SV, MV and DV.
fig. 3.1 0 Brightfleld (A, C) and darkfleld photomicrographs (B,D) of the floccular peduncle in K 244. Note the course of the labeled fibers through the floccular peduncle containing the cells of dorsal group y (pf+y) and the presence of fibers arching through the cerebellar nuclei (arrows in B). Labeled fibers ramifY among cells of dorsal group y (C). Bar In A= 500 ~m. inC= 33 ~m.
113
a 360-73
''~
~:)-j bp
'---'"-------__j .
~~c d 360-62
f 360-51
h 360·43
360-38
j 360.33 ~ ~.'· ... . ~·- \ -MV t
fig. 3.11 DraWings of labeled fibers and neurons in transverse sections through the flocculus and brain stem in case K 360 With an injection mainly located in FZJ, encroaching upon FZn. Note the efferent fibers in FCt, scattered fibers in FC2, and their termination in the superior (SV), medial (MV) and descending vestibular nuclei (DV).
114
f 301-107 lA >-
bo
bp
bo
g 301-123
;b
~L_'_'-" ___ '_" ____ '~~-"-"~_j
"
bo
(,~,,~ h 301-128 CO rb MV
M'l
""
' '
~ ~.Mv··. , , '!'H i 30H36
;b J 301-143
fig. 3.12 Drawings of labeled fibers and neurons in transverse sections through the flocculus and brain stem in case K 301 With an injection in FZn. Note the location of efferent fibers in the laterodorsal part of FCz and terminating in the medial vestibular nucleus (MV}.
ll5
a 305-53
~ !be/ v
' h 305-28
, PFLv 'J
j 305-16
co_~
rb • \ \.'1;', ~~ OV~
3.3.3 Conclusions
Retrograde labeling of Purkinje cells from HRP-injections in their
vestibular target nuclei confirmed the localization of the MY-projecting
cells in two zones, located rostral to, and interdigitating With two zones
projecting to SV (Yamamoto and Shimoyama. 1977). The axons of these
Purkinje cells collect in four bundles that correspond to the white matter
compartments FC1·FC4 delineated in the previous studies (Tan et a!.,
1989; chapters 1 and 2).
Anterograde WGA-HRP tracing combined With AChE histochemistry
demonstrates that Purkinje cells of climbing fiber zones project to
specific parts of the vestibular complex (fig. 3.17). Fir. and FZm mainly
project to the dorsolateral. central, and caudal parts of the SV, with
possible minor projections to DV. MV. and prepositus hypoglossi (PH).
FZn and FZtv project to MV, With possible minor projections to SV and
PH. Strong terminations were consistently found in the MV region
surrounding the acoustic stria. Most labeled fibers in DV seem to pass
through this nucleus on their way to MV rather than to terminate there.
The possible minor projections are likely due to spread to neighbouring
Purkinje cell zones.
Termination of fibers passing through group y could not be
definitely established. Ramifications of labeled fibers around neurons of
dorsal group y were observed in cases with injections of FZy and FZnJ, but
not with injections in the FZu zone. Terminations of arciform fibers from
FZ1 and FZu passing through the cerebellar nuclei and along Lpc cannot
be excluded. No arciform fibers were observed in the cases with
injections in FZm.
fig. 3.13 DraWings of labeled fibers and neurons in transverse sections through the flocculus and brain stem in case K 305 with an injection in FZn. Note the location of efferent fibers in FC2 terminating almost exclusively in the medial vestibular nucleus (MV).
117
Fine labeling was present in the dorsal cochlear nucleus in some of the
cases With injections in zones FZJ-FZm. but not in all of them. No
experiments With small injections in C2 and FZrv were available.
3.4 Discussion
The present study shows that the Purkinje cells of the flocculus are
partitioned in zones. Purkinje cell axons of each zone collect in a white
matter compartment and project to specific parts of the vestibular
nuclear complex. The data on pathways of Purkinje cell axons
demonstrated with anterograde tracing with WGA-HRP are supported by
the findings in our retrograde HRP experiments.
3.4.1 The retrograde and anterograde tracing methods used in this stw:iy.
Two methods have been used in this study to trace the efferent
pathways and connections of the flocculus: retrograde axonal transport of
HRP injected into the vestibular nuclei and anterograde axonal transport
of WGA-HRP injected into the flocculus.
Retrograde filling with HRP of Purkinje cells and their axons
demonstrated With DAB as a substrate has been used before to map the
corticonuclear and -vestibular projections (Voogd and Bigare. 1 980;
Voogd. 1989; Voogd et al.. 1987). Anterograde transport in vestibula
cerebellar mossy fibers that would have hindered the identification of
labeled axons as belonging to the retrogradely labeled Purkinje cells. was
never observed in these or in the present experiments.
WGA-HRP demonstrated with trimethyl benzidine (TMB) is a more
sensitive method. Moreover. WGA-HRP is transported both anterogradely
and retrogradely (Lechan et al., 1981: Trojanowski et al.. 1981). The
injections With ve:ry small amounts of WGA-HRP in the cortex of the
flocculus, used for anterograde tracing of the Purkinje cell axons, also
caused retrograde labeling of neurons in the vestibular nuclei bilaterally
and in the contralateral inferior olive. This raises the question whether
118
a 370-263 f 370-196
b 370-246
c 370.236
g 370-184
h 370-164
I 370-149
J 370.128
fig. 3.1 4 Drawings of labeled fibers and neurons in transverse sections through the flocculus and brain stem in case K 370 With an injection in FZm. Note the location of efferent fibers in FC3. their passage through the floccular peduncle and dorsal group y, and their termination in the superior vestibular nucleus (SV). Note the absence of fibers passing through the lateral cerebellar nucleus in comparison with cases K 360 and K 244 (figs. 3.9 and 3.!1).
119
some of the axonal labeling observed in our experiments could have been
due to retrograde transport in the axons of these labeled cells. This is
unlikely because few, if any vestibulocerebellar or olivocerebellar fibers
appear to be retrogradely labeled in these cases. Labeled olivocerebellar
axons were not observed in the olivary decussation or the restiform body.
No labeled fibers could be traced from labeled cells in the contralateral
vestibular nuclei to the flocculus. Moreover, the distribution of labeled
axons in the ipsilateral vestibular nuclei varies systematically with the
localization of the injection site, whereas the distribution of labeled cells
was roughly the same in all experiments (see Magras and Voogd (1985):
Thunnissen (1990) and Sato et al. (l983b) for experiments shoWing lack
of zonal differentiation in terminations of secondary vestibulocerebellar
mossy fibers). Anterograde transport of WGA-HRP of Purkinje cell axons
can, therefore. be used to study their distribution.
3.4.2 Zona! organization of Purkinje cells projecting to the media! and
superior vestibular nuclei and group y.
The overall distribution of the floccular Purkinje cell axons is
similar to earlier reports on the rabbit (Epema, 1990) and other species
(Voogd, 1964; Angaut and Broda!, 1967. cat; Haines. 1977, Galago; Langer
et ai .. l985b, macaque monkey). Epema (1990) noticed the two routes
followed by these fibers. One runs ventrally, through the floccular
peduncle, between the Lpc and the restiform body. Some of its fibers
proceed dorsal to LV to I.Owy's bundle near the lateral corner of the
fourth ventricle. The second route consists of arciform fibers. curving
dorsally and caudally through the lateral and anterior interposed
fig. 3.15 Darkfield and brightfield photomicrographs of adjacent AChEstained (A,C) and HRP-reacted sections (B,D) through the superior vestibular nucleus (A,B) and the floccular peduncle (C.D) in K 370 showing accumulation of terminating fibers in the dorsolateral part of SV (B). More caudally labeled fibers are almost exclusively located just dorsal to the restiform body in the floccular peduncle (D).
!20
121
cerebellar nuclei. Medial to Lpc and lateral to LV they rejoin the fibers of
the floccular peduncle. Arciform fibers. passing through the lateral
cerebellar nucleus were observed in cat and Galago. but not in the
macaque monkey (Langer et a!.. 1985b).
Our experiments with retrograde labeling of HRP show that
Purkinje cell axons of FZu and FZ!V, contained in the compartments FC2
and FC4, project to MV and those from FZ1 and FZm, located in FC 1 and
FCs project to SV. These experiments confirm and extend the
observations of Yamamoto and Shimoyama (1977) in the rabbit, especially
With respect to the presence of a medial FZyv zone and a corresponding
FC4 compartment, projecting to MV. Judging from the localization of the
injection sites. the distribution of the labeled Purkinje cell axons over the
white matter compartments and the retrograde labeling in the inferior
olive, most of the injections of WGA-HRP in the cerebellar cortex were
focussed on a single floccular zone with encroachment on neighbouring
zones in some experiments. No experiments With injections limited to
the FZN and C2 zones were available.
Our experiments demonstrate that FZr and FZm project strongly to
the SV, especially to its dorsolateral. central, and caudal parts. and that
FZu projects to the MV. Terminations in regions traversed by labeled
fibers such as group y, the lateral and anterior interposed cerebellar
nuclei, and DV are difficult to establish With certainty. In cases with
injections of FZI and FZn1 labeled fibers ramified among the cells of
dorsal group y, but no such ramifications were observed in cases with
injections in FZn. A few labeled fibers extended into nucleus prepositus
hypoglossi in cases with injections in FZu and FZm. Projections to the
dorsal cochlear nucleus preViously found by Epema (1990) were observed
in some, but not in all experiments With injections in FZI and FZUI·
Fibers from FZ1 and FZu travel both along the ventral route through the
floccular peduncle and group y, and the dorsal route, curving through the
lateral and anterior interposed cerebellar nuclei and the Lpc.
Terminations of Purkinje cell axons in these nuclei cannot be excluded.
122
I
K244 K355 K358 K350
93% 85% 82% 67% +------ ---------------I
7% 15% 18% 33% ---
250 81 78 86
in Nsections 49 12 18 14
-II -
K301 K340 K305
11% 15% --------------
89% 85% -------- . -46 35 -
28 25 ---------
. li-m r ill 370 K2271K
55% 8 5% ----
45% 1 5%
186 1 -------- -
' 19 1 L ___ L
50
8
fig. 3. I 6 Distribution of labeled cells over the caudal dorsal cap (cdc) and the rostral dorsal cap with the ventrolateral outgrowth (rdc + vlo) in cases with injections of WGA-HRP tn the floccular zones FZr-FZm.
Fibers from FZIII pass exclusively through the floccular peduncle. Our
obsenrations are in accordance With the results of De Zeeuw et al. (1992),
who traced axons from individual Purkinje cells of the flocculus with
biocytin. They found projections from Purkinje cells of FZr and FZm to
SV, and from FZn to parvicellular and magnocellular parts of MV. In
addition, they observed terminals from FZI and FZm Purkinje cell axons
in group y, and for FZI also in the ventral dentate nucleus. FZn did not
project to group y or the dentate nucleus.
Our findings are in good agreement with earlier studies {Yamamoto
and Shimoyama. 1977; Yamamoto. 1978; Balaban et al., 1981; Sa to et al.,
1982a, l982b). Our zone FZu and FZm correspond to the middle and
rostral zones of the flocculus of the cat (Sato et al., l982a) and the
monkey (Balaban et al., 1981). which also project to MV and SV,
respectively. By the absence of a projection to SV the caudal zone of the
cat flocculus differs from the FZ! zone of the rabbit (Sato et al., 1982b). It
is still possible that zones equivalent to FZ! and Cz exist in the caudal cat
flocculus, and that the injections of the lateral cerebellar nucleus and
123
group y labeled fibers of passage of FZ1 and C2 on their way to SV and the
posterior interposed nucleus. A caudal zone, which projects to SV and
corresponds to FZ1 was found in the monkey {Balaban et al., 1981).
The vestibular nuclei contain excitatory and inhibitory relay neurons
for the vestibula-ocular reflexes that convey input from semicircular
canals activated by head movements (Ito et al., 1973). Purkinje cells of
the flocculus innervate horiZontal and anterior canal relay cells, but do not
terminate on relay cells of the posterior canal (Ito et al., 1973, 1977,
1982a,b). Fibers from FZ1 and FZm terminate mainly in the dorsolateral,
central, and caudal parts of the SV. Excitatory anterior canal relay cells
that innervate contralateral motoneurons innervating the superior rectus
and the inferior oblique muscles, are located in dorsolateral SV.
Inhibitory anterior canal relay cells, which innervate inferior rectus and
superior oblique motoneurons of the ipsilateral oculomotor and trochlear
nerve nuclei, occupy the central part of SV (Highstein, 1973; Ito et al.,
1977; Yamamoto et al., 1978; Highstein and Reisine. 1979: Sato and
Kawasaki, 1990b; Labandeira-Garcia et al., 1991). Since floccular
stimulation inhibits the VOR pathway from the anterior canal (Ito et al.,
1977, 1982a,b) it is likely that it is conveyed through the FZr/FZm
projection to the central and dorsolateral SV. These vestibular neurons
are also active during vertical smooth pursuit eye movements (Chubb and
Fuchs, 1982: Chubb et al., 1984).
The heaviest projection of FZu is found in the parvicellular MV,
dorsomedial to the acoustic stria, and among the larger cells of MV,
ventrolateral to this bundle. Fewer fibers terminate in the most rostral
part of MV, ventral to SV, and in caudal parts of this nucleus. No fibers
were traced to LV or to the interstitial nucleus of the vestibular nerve.
Of the vestibule-ocular relay neurons located in l\1V, only horizontal
canal cells receive inhibitory projections from the flocculus {Ito et al.,
1973; 1977). Kawaguchi (1985) distinguished medial and lateral groups
of horizontal canal cells that received inhibition from the rabbit flocculus.
The medial group was located medial to the acoustic stria, and the lateral
group in the ventrolateral part of LV (!.e. in the magnocellular part of MV
124
in our terminology). Cells of both groups projected to the ipsilateral
medial rectus subdiVision of the oculomotor nucleus, but in addition the
medial group contained cells that presumably projected to the ipsilateral
abducens nucleus. These cells correspond to the inhibitory neurons
intercalated in the inhibitory vestibula-ocular pathway to the ipsilateral
abducens nucleus that is under inhibitory control of the flocculus
(Highstein, 1973; Ito et a! .. 1977; Sato et a! .. 1988; Sato and Kawasaki,
1991). The projection of the flocculus extends beyond the acoustic stria
into caudal MV and DV and. therefore, may include neurons with other
connections that the oculomotor nuclei (see also Epema, 1990).
The accumulation of fibers from FZI. FZn and FZm in the area of
group y indicates that some of these fibers may also terminate in this cell
group. Cytoarchitectonically it consists of dorsal and ventral parts
(Highstein and Reisine, 1979, cat; chapter 1. rabbit). The ventral group y
is formed by a thin layer of tightly packed small fusiform neurons
immediately dorsal to the restiform body. Neurons of ventral group y
receive primary vestibular afferents (Sa to and Kawasaki, 1987) from the
sacculus (Lorente de No, 1933; Gacek, ! 969; Carleton and Carpenter,
1984; Kevetter and Parachio, 1984). They do not respond to antidromic
stimulation from the oculomotor nuclei (Sato and Kawasaki, 1987). The
dorsal part of group y consists of larger, fusiform, and multipolar neurons
which gradually blend into the parvicellular part of the lateral cerebellar
nucleus. In the cat these neurons that receive polysynaptic input from
both VIIlth nerves are the primary target for floccular Purkinje cells in
the caudal zone (Sato and Kawasaki, 1987). Some of these target neurons
were also found in the adjacent ventral part of the LCN (Lpc). They
project to the caudal half of the contralateral oculomotor nucleus
(Highstein, 1971, 1973; Graybiel and Hartwieg, 1974; Highstein and
Reisine, 1979; Stanton, 1980; Carpenter and Cowie, 1985; Sato and
Kawasaki, 1987; Labandeira-Garcia et al.. 199!). This part of the
oculomotor nucleus harbours the motoneurons innervating the inferior
oblique muscle ipsilateral to the motoneuron and the contralateral
superior rectus muscle (Gacek, 1977; Naito eta! .. 1974; Akagi, 1978). In
our cases with injections of FZ1 and FZm labeled fibers ramified around
125
cells of dorsal group y that were not labeled retrogradely (figs 3.10; 3.15).
No such evidence was available for the projections of FZn.
3.4.3 Correspondence of cortico-vestibular with dimbing fiber zones in the
rabbitjloccu!us.
The zonal pattern described in the present study corresponds well
With the zonation of climbing fibers demonstrated with anterograde
WGA-HRP tracing in rabbit (chapter 2) and Phasaeolus leucagglutinin
transport in the rat (Ruigrok et al., 1992) and the autoradiographic study
of Gerrits and Voogd (1982) in the cat. In the previous chapter we
compared the climbing fiber projection to the flocculus with the
compartmentation in adjacent AChE-stained sections as a reference. We
found close correlations between particular subsets of olivary neurons and
certain white matter compartments. The cdc projected through
compartments FC2 and FC4 to floccular zones FZn and FZrv, while
rdc/vlo neurons projected through compartments FC 1 and FC 3 to
floccular zones FZI and FZm, Our observations on retrograde labeling of
cells of the dorsal cap and the ventrolateral outgrowth in experiments
with small injections of individual floccular zones generally support this
conclusion.
fig. 3.17 Diagram of the projections of the floccular zones FZJ-FZIV through compartments FC 1-FC4 to the vestibular nuclei.
126
3.4.4 Functional con.sideration.s.
The efferent floccule-vestibular projection forms part of the
neuronal circuitry that is the basis of oculomotor control (Ito et al.,
1973). Through these connections Purkinje cells in the flocculus can
inhibit the activlty of vestibular neurons (Highstein. 1971; Baker et al..
1972; Fukuda et al.. 1972; !to et al.. 1977). An imbalance of vestibular
discharge activities would result in eye muscle activity. Indeed, local
electrical stimulation in the flocculus elicits eye movements {Ron and
Robinson, 1973; Dufosse et al.. 1977; Sato and Kawasaki. 1984. 1990a,
l990b; Nagao et al., 1985; Belknap and Noda. 1987; Chapter 4). Some of
these studies indicated that a particular type of eye movement could only
be elicited by stimulating within a confined area of the flocculus (Sa to and
Kawasaki, 1984, 1990a. 1990b; Nagao et al., 1985; Simpson et al., 1989;
Chapter 4). Recent studies (Simpson et al., 1989, Van der Steen et al..
1991. Chapter 4) demonstrated that optimum responses were obtained
by stimulating the white matter of the flocculus. They verified the
stimulation sites histologically, compared them with the AChE
compartmentation, and found that in three of the five white matter
compartments (compartments FC1. FC2 and FC3) clear eye movements
could be elicited. The relatively small size of compartments Cz and FC4
made it difficult to be sure that the stimulation was confined to them and
no conclusion was reached about their relationship to eye movements
(see also chapter 4). Stimulation in compartments FC1 and FC3 resulted
in rotatory conjugate eye movements around axes in the horizontal plane:
the ipsilateral (left) eye moved counterclockwise around an axis at about
135' to the midsagittal plane, the contralateral (right) eye moved
clockwise around an axis at about 45° to the midsagittal plane. These
compartments flanked compartment FC2, in which electrical stimulation
resulted in horizontal eye movements (i.e. rotation around the vertical
axis) towards the stimulated side. The results of the present study and
that of Van der Steen et al. (1991, see also Chapter 4) indicate that each
floccular zone controls the activity of specific extra-ocular muscle pairs
127
through the compartmentaliZed flocculo-vestibulai- nucleus projection to
specific vestibula-ocular relay neurons in the vestibular nuclei.
!28
Chapler 4
Functional and anatomical organization of three-dimensional eye
mouements in rabbit cerebellar flocculus
4.1 Introduction
The flocculus of the cerebellum is part of the compensatory eye
movement control system that stabilizes gaze during involuntary head
rotations. Because head rotations can occur about any axis in space, signal
processing in the flocculus has to support eye rotations having three
degrees of freedom. Understanding this signal processing requires
detailed knowledge of how three-dimensional eye movements are
represented in the flocculus.
One approach to this question has been to use electrical stimulation
of the flocculus while recording eye movements or eye muscle actiVity
(rabbit; Ito et al.. 1977; Dufosse et al.. 1977; Ito et al.. 1982a.b: Nagao et
al .. 1985: monkey: Balaban and Watanabe. 1984: Belknap and Noda. 1987:
cat: Sato and Kawasaki, 1990a). Another approach has been to determine
the geometry of the floccular climbing fiber (CFJ signals of retinal image
slip (Maekawa and Simpson. 1973; Simpson and Alley. 1974; Simpson et
al.. 1981: Leonard et al.. 1988; Graf et al.. 1988: Kana et al.. 1990:
Kusunoki et al., 1990). From the electrical stimulation studies, a
topographical representation in the form of zones emerged from the
distribution of sites where microstimulation evoked different classes of
eye movements or influenced different vestibula-ocular reflex (VOR)
pathways. A zonal representation is also indicated from the CF studies
showing that the different classes of CFs that signal retinal slip in
reference to specific axes of visual world rotation arise from different
parts of the dorsal cap and ventrolateral outgrowth of the inferior olive.
When this relation is combined with the general anatomical and
physiological finding (Voogd. 1969: Groenewegen and Voogd. 1977:
Voogd and Bigare. 1980; Broda! and Kawamura. 1980; Gerrits and Voogd.
129
1982: Oscarsson. 1969. 1979) that the inferior olive can be subdiVided
such that each subdivision's terminal field in the cerebellar cortex has the
form of a parasagittal zone, then the importance of a zonal configuration
in the floccular representation of eye movements is further apparent.
Even so, the reported localization and extent of these zones Within or
between species has not been consistent. The variable and sometimes
conflicting results have arisen for at least two reasons. First. the
techniques used to measure the evoked eye movements did not permit
precise, dynamic three-dimensional measurements. Second, no
independent anatomical delineation of the zones was available for
comparison with the physiology on an animal by animal basis. To
overcome the first limitation. three-dimensional search coils can be used
to obtain the required measurements (Van der Steen and Collewijn.
1984). An opportunity to overcome the second limitation has been
provided by the finding of Hess and Voogd (1986) that
acetylcholinesterase (AChE) histochemistry can be used to reveal distinct
compartments in the cerebellar white matter. With AChE histochemistry
five compartments have been demonstrated within the floccular white
matter of the rabbit (Tan et al., 1989; Van der Steen et al., 1989; see
chapter 1).
This paper reports that in the rabbit implanted with three
dimensional search coils only a small number of different classes of short
latency eye movements are evoked by electrical stimulation of the
floccular white matter and that they are differentially associated with
anatomically distinguishable compartments revealed by AChE histo
chemistry. Furthermore, these eye rotations are organized in a coordinate
system like that of the visual CFs. In view of the similarity of the
orientation of the preferred axes of the CFs and the rotation axes of the
extra-ocular muscle pairs, a relatively simple geometrical relationship is
suggested between each class of CFs and the eye rotations produced by
the activation of the Purkinje cells upon which that class of CFs synapse.
Some of the results were presented previously in brief (Van der Steen et
al .. 1989. 1991).
130
4.2 Materials and methods
4.2.1 Eye movement recordings.
Eye movements in response to electrical microstimulation in the
flocculus were recorded tn 44 awake, pigmented rabbits. The
three-dimensional movements of both eyes were recorded with the
scleral search co!l technique (Van der Steen and Collewijn. 1984). One
week before the experiment two search coils, one horizontal on top of the
superior rectus muscle and one vertical, parallel to the limbus, were
implanted using general anaesthesia (initial intramuscular dose: 32 mg/kg
ketamine. 0.6 mg/kg acepromazine, 5 mg/kg xylazine; supplement each
45 min.: 18 mg/kg ketamine, 0.32 mg/kg acepromazine. 4 mg/kg
xylazine) and aseptic techniques. The vertical axis (horizontal)
component of eye rotation was measured With the vertical coil using the
phase-angle detection technique (Collewijn,l977; Van der Steen and
Collewijn. 1984), while the rotation components about the 45' and 135'
axes (see below) were measured with the horizontal coil using the
amplitude detection technique (Robinson, 1963). The vertical coils had
absolute calibration, whereas the horizontal coils were calibrated prior to
implantation. The three eye position components for each eye were
recorded on a polygraph and stored on magnetic tape. To generate eye
position plots. the components were digitized off-line on a minicomputer
(DEC, PDPll/73) at a sampling frequency of 125 Hz with a resolution of
10 seconds of arc. The latencies of the eye movement components were
taken from selected trials throughout the experimental series. The
latencies were determined from eye position signals digitized at a rate of
500 Hz by a CED-1401 data-acquisition system (Cambridge Electronic
Design) connected to a personal computer. Each axial component was
displayed at its optimal resolution and its latency from stimulus onset was
determined from the display by judging where the eye displacement first
exceeded the noise level. Because of the variations in the shape of the
initial eye movements following electrical stimulation, a visual inspection
procedure was chosen rather than an automatic fitting procedure.
131
VA
----3 0 45 cc
fig. 4.1 The orthogonal coordinate system used in measuring the angular displacements of the eyes. For explanation see text.
4.2.2 The coordinate system for measuring the eye movements.
The orthogonal coordinate system used to measure the eye
rotations is shown in fig. 4.1. It consisted of a vertical axis and two
horizontal axes oriented at 45° and 135° azimuth. The oo azimuth
reference was defined as rostral in the midsagittal plane and the
azimuthal coordinate was taken to increase to each side of the reference
direction. The three components of the rotation of each eye were named
after their respective axis of measurement (i.e. vertical axis (VA), 45°
axis, and 135' axis). The sense of rotation was defined as left and right for
the vertical axis and clockwise (CW) and counterclockwise (CCW) for the
132
45° and 135° axes according to how the resPective eye movement
components would appear to an observer looking along each axis of
rotation towards the rabbit's eye. With these conventions, the 135° axis of
one eye is oriented parallel to the 45° axis of the other eye, and a CW
rotation of one eye is conjugate. with respect to sense of rotation to a CCW
rotation of the contralateral eye.
4.2.3 Electrophysiology.
Several days before the expertment the rabbit was anaesthetized (as
above) and under aseptic conditions a craniectomy was made to expose
the paramedian lobe on each side of the cerebellum. A small chamber of
dental acrylic was constructed around each opening in the skull. The dura
was covered with an antibiotic gel and a thin sheet of silicon rubber, and
the chamber was filled with bonewax and paraffin. On the day of the
experiment, the rabbit was comfortably restrained in a hammock with the
head held with the nasal bone at 57' to the hortzontal. The dura was
opened under local anaesthesia and a recording microelectrode (glass or
metal) was lowered into the cerebellum. The flocculus was identified by
monitortng the modulation of the Purklnje cell CF act1v1ty in response to a
handheld moving visual pattern (Simpson and Alley, 1974; Graf et al ..
1988). With the animal in darkness, the flocculus was stimulated along
the electrode track at 200 ~ intervals with electrical pulses (200 Hz,
2-20 ~ negative polartty, 0.2 msec pulse width, 1 sec train duration). A
stimulation site was regarded as an effective site when electrical
stimulation, not exceeding 20 ~ produced an eye movement with an
amplitude of at least 0.5° about at least one of the siX axes of rotation.
When metal electrodes were used, selected sites effective for evoking eye
movements were marked by small electrical lesions (5 ~. 5 sec duration
DC current. tip positive). Experiments were carried out using both the
right and left flocculi, but for clarity of presentation all results are
represented in reference to stimulation of the left flocculus.
133
4.2.4 Histology.
See chapter 2.2.
4.3 Results
4.3.1 Classification of eye movement responses.
Prior to the availability of the AChE staining showing white matter
compartments, experiments were conducted in 26 rabbits to determine
the characteristics of the three-dimensional eye movements evoked by
electrical stimulation of the flocculus. In this series of experiments about
90% of the flocculus was explored; only the most rostral and the most
caudal parts were not included. Eye movements were more readily
evoked by stimulation in the deep granular layer and in the white matter
than from the molecular and Purk.inje cell layers. Typically the responses
consisted of slow eye movements with an amplitude varying between 1°
and 8°. On occasion, slow eye movements with amplitudes up to 20° were
observed and these larger movements were sometimes interrupted by
resetting saccades. The onset of the prominent components of the eye
movement responses usually occurred between 8-48 ms after stimulus
onset. Peak speed occurred 100-200 msec after movement onset and
reached a maximum value of 20°/S. After cessation of the stimulus the
eyes usually returned within 4-10 seconds to the position held before the
stimulus began. In a few instances. however, the eyes remained in an
eccentric position and successive l sec periods of stimulation then
produced successively smaller displacements. The eye movements were
classified according to the largest component of the response. Of the 12
possible classes (2 eyes x 3 axes x 2 senses of rotation), two of the classes
accounted for 78% of the responses.
4.3.2 135° axis eye movement responses.
For 59% (82/140) of the effective stimulation sites. the largest
134
response component was a CCW rotation of the left (ipsilateral) eye about its 135° axis. Three examples of this class of eye movements, each from a
different animal. are shown In fig. 4.2. The 135' axis class showed three
variations: In about 50% (42/82) of the cases the response was conjugate
in that the largest response component for the contralateral (right) eye
was a CW rotation about Its 45' axis. The amplitude of this component was
always smaller than the conjugate component about the 135' axis of the
Ipsilateral (left) eye. In 13% (10/82) of the cases. the conjugate
component of the contralateral eye was virtually absent. The third
variation (37%, 30/82) consisted of cases where. In addition to the CCW
135' axis response of the Ipsilateral (left) eye (which was the largest of all
components). the CW component of the contralateral (right) eye about Its
45' axis was accompanied by an equally large or larger CCW component
about the 135' axis (see e.g. fig. 4. 7 A). For each of the three variations of
the 135° axis class, disjunctive rotations of both eyes about the vertical axis were sometimes present, as shown by the VA traces in fig. 4.2.
4.3.3 Vertical axis eye movement responses.
For 19% (26/140) of the effective stimulation sites the largest
response component was an abduction of the ipsilateral eye (a leftward
rotation of the left eye about Its vertical axis). Fig. 4.3 shows three
examples, each from a different animal, in which the ipsilateral VA component was the largest of all components. The VA component of the
contralateral eye was conjugate in direction, but showed considerable
variation In magnitude. In the example of fig. 4.3 A the evoked eye
movement was virtually restricted to the ipsilateral eye. In other cases
(fig. 4.3 B-C) stimulation resulted In rotation of both eyes about the
vertical axis, although the VA component of the contralateral eye was always smaller than that of the ipsilateral eye. In addition to the major
component of rotation about the vertical axis, smaller rotations about the
45° and 135° axes sometimes occurred.
135
A 8 c 0 Stimulus
Right (Central Eye
VAJ: ~ Jccw e "~ ~ ~ cw
Jccw T N
135------ ----------cw
Lett (lpsil Eye
VAr------ ~
r \/V ~ e! 135~ ccw N T
r "~ ccw 4 sec [ 5"
fig. 4.2 Examples from three different animals of the class of evoked eye movements for which the largest component was a CCW rotation of the ipsilateral (left) eye about its 135' axis. In these examples (A-C) the largest component of the contralateral (right) eye movement was a conjugate (CW) rotation about its 45° axis. For this class of eye movements, an accompanying disjunctive (vergence) movement about the vertical axis was often present. In this and other similar figures, the three traces for each eye show the angular displacement about the indicated axes (L, leftward; R, rightward; CW, clockwise; CCW, counterclockwise). A downward deflection indicates a leftward rotation of each eye about its vertical axis (VA). a clockwise rotation of the right (contralateral) eye about its 45° and 135° axes, and a counterclockwise rotation of the left (ipsilateral) eye about its 45° and 135° axes. The periods of electrical stimulation are indicated by the solid blocks at the top. The drawings to the right of the eye position traces in this (D) and other similar figures show how the net rotation of the eye would appear if a cross had been placed on the cornea {dashed cross: initial position. solid cross: position at the end of the stimulation. T, temporal; N, nasal).
136
A 8 c 0 Stimulus
VAJ: Right IContral Eye
~~ ~ JCCW e 45
CW
Jcc~ T N
135 --------cw Left llpsil Eye
VAlR~ ~ ~ JL
@ Jew 135 ~ ~ ccw
N T
r ~ 45 ccw ~
[ 5' 4 sec
fig. 4.3 Examples from three different animals of the class of evoked eye movements for which the largest component was an abduction of the ipsilateral (left) eye. In the left panel (A) the response was virtually restricted to the ipsilateral eye. whereas in panels B and C tbe responses were bilateral. For explanation of symbols and abbreviations see fig. 4.2.
4.3.4 Other response types.
Although the majority of the responses fell into one of the two
classes described above, In the remaining 22% (32/140) of the cases
other types of responses were recorded. These responses comprised, in
part (10/140}, cases in which the largest component was a CCW rotation
about the contralateral (right) 135' axis In combination with a smaller CW
rotation about the contralateral (right) 45' axis. In other cases (16/140) a
CW 135' axis rotation of the ipsilateral (left) eye was the largest
component.
137
The remaining 6 cases consisted of vergence eye movements for which
the disjunctive contralateral VA component was the largest component.
4.3.5 Latencies of the response components.
A further distinction between the different types of eye movement
responses was made on the basis of the latency of each of the
components. For this purpose. examples were drawn from throughout the
experimental series. Using the larger amplitude responses facilitated the
latency measurements. which revealed short and long latencies among
the different components.
The CCW 135' axis and abducting ipsilateral components. which
defined the two main classes. all had short latencies. Of these, the
latencies of the 135° class were the shortest and had the sharpest onset.
A typical example of the onset of both the ipsilateral CCW 135' and
contralateral CW 45° axis components is shown with an expanded time
and position scale in fig. 4.4 C. Fig. 4.4 A shows the latency distribution
for the population of analyzed 135° class responses. The mean latency of
the CCW rotation of the ipsilateral (left) eye about its 135' axis was 15
msec (S.D. ± 11 msec, n=65}. When both eyes moved there was also a
measurable CVV rotation of the contralateral (right) eye about its 45° axis.
The mean latency of this response was 28 msec (S.D. ± 18 msec. n=35).
This difference in the onsets of the ipsilateral CCW 135' and contralateral
CW 45° axis responses was not significant {p=0.74, Kolmogorov-Smirnov
test).
The latencies of the VA components of the VA class responses were
also short (fig. 4.4 B-C). The mean latency of the ipsilateral abducting VA
response was 27 msec (S.D. ± 15 msec, n=27), while the mean latency of
the contralateral adducting VA component was 49 msec {S.D. ± 25 msec,
n=l9). These latency differences were not significant (p=0.5.
Kolmogorov-Smirnov test).
The latencies of the other relatively large response components
were also measured. In one of the three variations of the 135° axis class,
stimulation evoked a CW rotation of the contralateral {right) eye about its
138
A
" " " • ~ " •
" " • " " 3'
"
c
,g 0.60
~ 0.40
0.20
Eye Movement Response Latencies
B ipsi 135" and contra 45" -lpsi 135" Ji'lE3'3 Contra 45"
0 20 40 60 80 100 120 140 160 160 200 Latency· (msecJ
Contra 45"
lpsi 135"
too zoo 3oo 400 sao Boo Time (msecl
ipsi VA and contra VA lllilllli lpsi VA illili2J Contra VA
:: f "
~I ~~-~~ I D
1.00
0.80
0.60
0.40
0.20
0 20 40 60 80 100 120 140 100 160 200
Latency (msecJ
Ccntrn VA
IpS! VA
o.oo ~L-~-~-~-~-_._:,___ 0 100 200 300 400 500 600
Time (msec)
fig. 4.4 Latency histograms (top panels) and examples (lower panels) for the predominant component and the conjugate component for the 135° axis (A and C) and VA axis (B and D) classes of evoked eye movements. For each class the ipsilateral and the contralateral conjugate components are indicated by solid and hatched bars, respectively. In the histograms the number of ipsilateral cases is greater than the number of contralateral cases because in some instances there was no accompanying response component for the contralateral conjugate axis.The latencies were grouped in 5 msec bins and two bins, one for the Ipsilateral component and one for the contralateral component, are displayed for each latency interval. The top trace in C and D shows the onset of the stimulation period. Further explanation in the text.
45' axis along with a comparably sized CCW rotation about its 135' axis.
The latencies of the CCW rotation about the contralateral (rtght) 135' axis
and the CW rotation about the contralateral (right) 45° axis were clearly
different (fig. 4.5 A, C). The component about the 135' axis had a
139
A
c
Eye Movement Response Latencies
" " 30
contra 135' and contra 45' lllilllllll Contra 135" Mil Contra 45'
a 20 40 ao ao 100 120 140 1ao Teo 200 Latency rmsecJ
Contra 135"
Contra 4SO
o.oo L-~-~-~---~-~ 100 200 300 400 500 600
Time rmsecl
8 .. A vergence - 1ps1 V and contra VA
D
llllill lpsi VA M Contra VA
;:I 30
" 20
:~ r I . t 1!1!! Mll!~na,l~~l~ IM 1 u 1,1
0 20 40 60 60 100 120 140 160 180 200
Latency (msecl
'" w .-r""
::r::'~;;; ::: ~ 0.00 ~---------~-~
100 200 300 400 50\l 600
Time (msecl
fig. 4.5 Latency histograms (top panels) and examples (lower panels) for evoked eye movements for which (A and C) rotation of the contralateral eye occurred with nearly equal amplitude components about the 45' and 135' axes, or for which (B and D) vergence occurred with adduction of both the ipsilateral and contralateral eye about the vertical axis. The format for the latency histograms In A and B Is similar to that used for Fig. 4 A and B. For the example InC, the latency of the contralateral CCW 135' axis component was 90 msec longer than that of the contralateral CW 45° axis component. For the example In D. the latency of the ipsilateral adducting component was 125 msec longer than that of the contralateral adducting component.
substantially longer mean latency (121 msec, S.D.± 46 msec, n=22J than
that of either the accompanying CW 45' axis component (27 msec. S.D. ± 16 msec, n=22) or the ipsilateral accompanying CCW 135° axis
component (14 msec, S.D. ± 7 msec, n= 22). A.s can be appreciated from
the latency distribution shown in Fig. 4.5 A, the latency differences
between the contralateral CW 45' and the contralateral CCW 135' axis
140
components were statistically highly significant (p<0.001, Kolmogorov
Smirnov). It should be noted that for this variation of the 135° axis class
the movements of the two eyes were far from conjugate. The net
response of the contralateral eye was approximately "vertically" upward,
whereas the response of the ipsilateral eye was predominantly a CCW
rotation about its 135° axis.
Some eye movements of the 135° axis class included an adduction
of both the ipsi- and contralateral eyes about the vertical axis, producing a
disconjugate (vergence) movement. The latencies of the accompanying
ipsilateral CCW 135° axis and contralateral CW 45° axis components were
short, as described above. However, the latencies of the disconjugate VA
components of the two eyes differed considerably (fig. 4.5 B. D). The
onset of the ipsilateral VA response in the adducting direction was
considerably later (142 ms, S.D. ± 53 msec, n=29) than that of the
contralateral adduction (59 msec, S.D. ± 20 msec. n=l3), The two
latency distributions, shown In fig. 4.5 B, were significantly different (p<
0.0!, Kolmogorov-Smirnov).
In those instances in which the ipsilateral eye rotated about the
135° axis in the CW sense rather than in the far more commonly found
CCW sense. the latencies of the CW component were significantly longer
(106 msec, S.D. ± 16 msec, n=l3, p<0.001, Kolmogorow-Smirnov).
4.3.6 AnatomicaHy delineated compartments.
In the course of the experiments. the AChE histochemistry
technique Introduced by Hess and Voogd (1986) became available for use
in the rabbit. With this technique, the floccular white matter was seen to
be subdivided into five compartments separated by darkly stained raphes
(Tan et al., 1989; Van der Steen et al., 1989; chapter l). Following the
convention described in chapter 1, the compartments were labeled from
lateral to medial as Cz, FC1, FC2. FC3 and FC4. Fig. 4,6 A and B show
AChE stained transverse sections through the rostral and caudal thirds of
the flocculus, respectively. A complete series of transverse sections taken
at 80 IJ.m intervals through the entire flocculus of one animal was digitized
141
c Rostral
Caudal
fig. 4.6 Anatomically delineated compartments in the floccular white matter. A and B. AChE-stained transverse sections, showing the different compartments separated by darker staining raphes. A is from the rostral third of the left flocculus, and from lateral to medial the compartments are labeled (FC)l, 2. 3, and 4. B !s from the caudal third of the left flocculus. Because the compartments run obliquely from caudomedial to rostrolateral. compartment 4, which is the most medial one. is not yet clearly visible in this relatively caudal section, whereas C2 is present only in the most lateral part of the white matter. The scale bar in B represents l mm and applies also to A. bp: brachium pontls. The full extent of the floccular white matter compartments. in C, is based on a threedimensional reconstruction of the white matter. The reconstructed floccular white matter is shown as a projection onto a horizontal plane below the flocculus with the observer looking down on the plane.
142
and a three-dimensional reconstruction of the white matter was made
using a graphical computer program (Hillman et a!, 1990; Mahoney et al.,
1990).The reconstruction (fig. 4.6 C) shows the five compartments
forming a set of strips running obliquely in a caudomedial to rostrolateral
direction.
4.3. 7 Classes of eye movements in relation to anatomical compartments.
The relation between the different classes of three-dimensional eye
movement responses and the anatomically distinguishable compartments
was elucidated in a second series of experiments performed on 18 awake,
pigmented rabbits. In this series, the electrical stimulation and eye
movement recordings were combined with AChE histochemistry. The eye
movements were evoked by electrical stimulation with tungsten
microelectrodes and small electrolytic lesions were placed at sites
producing different classes of eye movements. In the AChE-stained serial
sections the lesions and electrode tracks were located With respect to
the anatomically distinguishable compartments. With this technique a
clear relation was found between compartments FC1, FC2 and FC3 and the
different classes of eye movements. Because the most caudolateral and
most rostromedial compartments (C2 and FC4, respectively) are
comparatively small, it was not possible to determine the eye movements
evoked by stimulating only the fibers in the white matter or granular layer
within these two compartments.
Eye movements whose largest component was a CCW rotation about
the ipsilateral (left) 135° axis were evoked by stimulation of the most
lateral of the three large compartments (FC 1). An example with two
stimulation sites in FC1 is shown in fig. 4.7. In this particular case, the
electrode passed through folium p (Yamamoto and Shimoyama, 1977;
Yamamoto, 1978) in addition to the more ventral part of FC I· At both
lesion sites (A and B) the largest component of the eye movement
response was a CCW rotation about the 135' axis of the ipsilateral (left)
eye, accompanied by a smaller CW rotation of the contralateral eye (right)
143
A Stimulus
Right !Contra) Eye
r~
::J~cw~~ cw
Jccw r--
135 ~ cw Lelt (lpsil Eye
135Jcw \ ~ ccw ~ \__..-.----~
N T
]
CW 45 ~ ccw
4 sec
B
vAJ:
Jccw 45 cw
Stimulus
Right !Contra) Eye
Jccw-'35 cw
Left (lpsil Eye
VA]~~ 135]cw \ /\
ccw '-" v 45]cw ~
CCW
@ T N
~ N T
fig. 4. 7 Eye movements evoked by electrical stimulation in compartment FC1. The drawing in the center panel depicts a transverse section of the flocculus; white matter compartments FC 1, FCz and FC3 are indicated by large, medium and small dots, respectively. The eye movements in A resulted from stimulation in the ventral part of compartment FC1 {marking lesion indicated by large arrow). The responses in B were evoked by stimulation in the extension of compartment FC 1 into folium p (marking lesion indicated by small arrow). In both cases. the largest response component was a CON rotation of the ipsilateral (left) eye about its 135° axis.
about its 45° axis. These movements were accompanied by vergence
components about the vertical axis and. as shown clearly In fig. 4. 7 A. also
by a CCW rotation about the contralateral 135° axis.
Fig. 4.8 illustrates the eye movements evoked by stimulation of
compartment FC3. The eye movements produced by stimulation of FC3
were on aggregate indistinguishable from those produced by stimulation
144
8 Stimulus
Right (Contral Eye
vAJ:
JCCW-------------
45 cw
Jccw
135 cw
Left Upsil Eye
~ T N
VAJ:
135]CW~e CCW N T
45JCW --------CCW ~
4 sec L5o
fig. 4.8 Eye movements evoked by electrical stimulation in compartment FC 3. The largest response component was a CCW rotation of the ipsilateral (left) eye about its 135° axis (B). Also note tbe small disjunctive rotations of the eyes about the vertical axis. The drawing of the transverse section in A shows the marking lesion (arrow) at the stimulation site in compartment FC3 (small dots), which at this relatively rostral level comprises the majority of the white matter. The laterally located compartment FC2 is indicated by medium dots and the thin. medially located compartment FC4 is undotted.
of FC1: the largest component was a CCW rotation about the 135° axis of
the ipsilateral (left) eye. The variations of the 135° axis class found for
compartment FC 1 stimulation were also found for FC3 stimulation,
145
Stimulus
Right iContraJ Eye
VAJ:
45 ----]
CCW
CW
]
CCW 135 cw _/-~-
Lett (lpsil Eye
13s]cw 'vi , CCW v
Jew
45 ccw~
~ T N
~ 1 mm
N T
10']
4 sec
Stimulus
Right (Contral Eye
VAJ:
]ccw
45 -----CW ~
T N
]
CCW 135 --------cw
Left (lpsil Eye
Jew
135 ccw w
N T
Is· Jew
45 ccw
fig, 4. 9 Example of the sharp change in the class of eye movement responses occurring when the stimulating electrode passed from one compartment to the other along the direction indicated by the arrow. The center panel shows the stimulation sites in compartment FC3 (filled square, fine dots) and in compartment FC2 (filled circle, medium dots). The eye movements produced at the two locations are shown in the two bordering panels. A) For the eye movements evoked from compartment FC3 the largest component was a CCW rotation of the ipsilateral (left) eye about its 135' axis. Note that the vertical axis rotation is disjunctive and that the ipsilateral VA component is an adducting (rightward) rotation. B) Stimulation 200 ~m deeper than in (A) and in compartment FCz produced VA class eye movements. The largest component of the ipsilateral eye movement was an abducting (leftward} rotation.
including disjunctive VA (vergence) movements.
The raphes form distinct anatomical boundaries between the
different white matter compartments. If these compartments also
represent functional subunits. then one would expect to find marked
146
changes in the eye movement responses when the stimulating electrode
passed from one compartment to the other. The presence of such a
relation between anatomical and functional compartmentation is shown in
figs. 4.9 and 4.10. Fig. 4.9 illustrates a case in which the electrode
passed through the white matter of FC3 and then entered the white
matter of FC2. as depicted in the central panel. The two different classes
of eye movements shown in fig. 4.9 A and B were evoked from stimulation
sites only 200 ~J-m apart. Fig. 4.9 A shows the eye movements obtained
with the electrode tip located in FC3 (black square in central panel). The
dominant rotation component was a CCW rotation about the ipsilateral
(left} 135° axis. accompanied by some vergence about the vertical axis.
When the electrode was advanced 200 11m into FCz (filled circle in
central panel) a sharp transition of the eye movement responses
occurred. The dominant response component became an abduction of the
ipsilateral eye (fig. 4.9 BJ.
Stimulation of FC2 invariably resulted in eye movements whose
largest component was an abduction of the ipsilateral eye, as illustrated in
figs. 4.9 B and 4.10. In the case illustrated in fig. 4.10. two electrode
tracks were made at the same rostrocaudal level. The more lateral track
passed into FC 1, while the more medial track passed into FC2.
Stimulation in FC 1 evoked eye rotations whose largest component was a
CCW rotation about the ipsilateral (left) 135' axis (fig. 4.10 A). and
accompanying disjunctive movements about the vertical axis. In contrast,
stimulation in the adjacent compartment (FC2) evoked eye rotations
whose largest component was an abduction of the ipsilateral eye (fig. 4.10
B). The sharp transitions in the responses evoked by electrical
stimulation in relation to the floccular white matter compartments
indicate that they are !n correspondence to the physiologically
distinguishable classes of eye movements.
147
A Stimulus
Right (Contra) Eye
vAJ:~
45Jcc~~ ~ T N
Jccw 135~
ew
Left Opsil Eye
vAJ:~
jew 135 Y'\. ~ ....----'\.,.
ecw 1 '-../ ·~
lew 45Jce~
4 sec ~.,
<@ N T
B Stimulus
Right (Contra) Eye
VAJ:~
45J:~-w ~-~ Jeew
135 ----------ew
Left (lpsil Eye
vAJ:~ 135lcw~ ;ffi
'Jcew ~ N
Jew 45~
ecw ~ 4 sec lso
148
4.4 Discussion
This study provides a direct demonstration of a correspondence
between anatomically identified floccular zones and different patterns of
evoked eye movements. In previous studies using electrical stimulation of
the flocculus, a correspondence could only be inferred because no
independent anatomical identification of floccular zones was available for
the animals actually used in the stimulation experiments. In those
studies, the anatomically based zonal structure of the flocculus was
compiled from a number of different animals using retrograde HRP
labeling techniques to delineate differential projections of floccular
Purkinje cells to the vestibular nuclear complex (Yamamoto and
Shimoyama, 1977; Yamamoto. 1978 ; Sato eta!.. 1982a,b; Balaban et al.,
1981). In the present study, however, the floccular zones could be
determined in each stimulated animal by staining for AChE.
Previous investigators have pointed out that the wide variations
among animals makes it difficult to use standardized maps (Nagao et a!..
1985; Kusunoki et a!., 1990). Attempts to superpose results from a
number of animals have led to the impression that the zones are more
interdigitated than is seen using AChE staining. With an independent
anatomical marker of floccular subdivisions available for each animal, a
more accurate and reliable comparison between physiological and
anatomical maps of the rabbit flocculus can now be made. To this end, we
will first discuss the present findings on the classes of evoked eye
movements and compare them to previous findings in the rabbit. Second,
we will discuss the correspondence between the present anatomical and
fig. 4.1 0 Association of different classes of eye movements with different anatomically delineated floccular white matter compartments. Top panel: AChE stained transverse section showing electrolytic marking lesions in two adjacent compartments of the floccular white matter. Electrical stimulation in the white matter of compartment FC 1 at the site marked by the more lateral lesion (a) evoked 135° class eye movements, as shown in A. In this particular case. the accompanying adduction of the ipsilateral eye was unusually large. Stimulation in compartment FC2 at the site marked by the more medial lesion (b) evoked VA class eye movements as shown in B. The scale bar represents 1 mm.
149
physiological floccular maps and compare them to previous attempts to
chart the flocculus structurally and functionally.
4.4.1 Classes of evoked eye movements.
The sense of rotation and the orientation of the rotation axes of the
short latency classes of evoked eye movements are in line with the effects
of flocculus stimulation in rabbit on the various semicircular canal
vestibula-ocular reflex (VOR) pathways (Ito et al.. 1973, 1977, 1982b).
Floccular Purk.inje cells monosynaptically inhibit vestibular nuclei neurons
that are part of the three-neuron arc of the VOR (e.g. Ito et al., 1970;
Fukuda et al., 1972; Baker et al.,1972; Highstein, 1973; Kawaguchi,
1985; Sato and Kawasaki, 1987, 1990b; Sato et al.. 1988). but only some
of these VOR pathways are directly influenced by the flocculus. The
specificity of the inhibition for particular pathways was shown by
determining whether electrical stimulation of the flocculus depressed
indiVidual canal-ocular reflexes. These experiments demonstrated that
the rabbit flocculus influences six specific canal-ocular pathways (Ito et
al., 1973, !977, l982a,b). Two of these pathways link the horizontal
canal with the ipsilateral medial and lateral recti muscles. Increased
actiVity of the related floccular Purkinje cells would result in increased
activity of the lateral rectus muscle and decreased activity of the medial
rectus muscle, causing abduction of the ipsilateral eye. In our
experiments. such eye movements formed one class of the short latency
responses. The four other canal pathways under direct floccular control
arise from the anterior canal. They excite the ipsilateral superior rectus
and the contralateral inferior oblique muscles and inhibit their
antagonists. the ipsilateral inferior rectus and contralateral superior
oblique muscles. The eye movements expected as a result of increased
actiVity of the Purkinje cells influencing the anterior canal pathways
would be, for a rabbit. a combination of depression, intorsion and
abduction of the ipsilateral eye and a combination of intorsion, elevation
and adduction of the contralateral eye. In our experiments such eye
movements constituted the majority of the short latency responses.
!50
The short latency eye movements produced· by floccular stimulation
in our study are in agreement with the results of Ito et al. (1977) who
showed, using electromyographic and muscle tension measurements. that
floccular stimulation produced depression of the excitatory and inhibitory
three neuron arc pathways originating from only the horizontal and
anterior canals. The posterior canal pathways were not inhibited by
floccular stimulation. In agreement with this finding we never observed
an eye rotation that had as its main component a rotation about the
ipsilateral 45° axis, which would have implicated the posterior canal
pathways. Ito et al. (1977) proposed four separate groups of flocculus
inhibited neurons in the anterior canal pathways: two groups of excitatory
neurons, one projecting to ipsilateral superior rectus motoneurons and
the other projecting to contralateral inferior oblique muscle
motoneurons; and two groups of inhibitory neurons, one projecting to
ipsilateral inferior rectus motoneurons and the other projecting to
contralateral superior oblique motoneurons. Graf et al. (1983) showed,
however. that rabbit posterior canal VOR relay neurons branch to both
oblique and vertical rectus motor pools. If a similar branching pattern
applies to anterior canal VOR relay neurons,as was suggested for cat by
Sate and Kawasaki (1991), then the number of distinct groups of anterior
canal pathway neurons receMng Inhibitory input from the flocculus would
be two rather than four. On the other hand, in our experiments in rabbit
quantitative differences were found between the amplitude of the
rotations of the ipsilateral and contralateral eyes. Often movement of the
contralateral eye was virtually absent, which indicates that the projection
pattern of floccular receiving neurons is probably not uniform and that
some flocculus receiving neurons in the rabbit vestibular nucleus project
to the motoneurons of only one extra -ocular muscle.
Although there is general agreement between the predicted eye
movements and those evoked in the present study, there are some
important differences With previous reports on eye movements evoked by
electrical stimulation of the rabbit flocculus. In experiments by Dufosse et
al. (1977) and Nagao et al. (!985) the eye movements were classified as
horizontal, rotatory, and vertical. The eye rotation axes of the horizontal
151
and rotatory classes, as defined in these earlier investigations, are
consistent with the eye rotations expected from combining the
electromyographic studies of Ito et al. (1973, 1977. 1982a.b) with the
measurements of the pulling directions of the individual extra-ocular
muscles in the rabbit (Graf and Simpson, 1981: Simpson and Graf. 1985).
The VA class of eye movement responses found in the present study is
consistent With the previously defined horizontal class (Dufosse et al.,
1977) and can be explained by the known influence of the flocculus on
the lateral and medial recti muscle activity.
In the present study, the most common response was a CJ::IN
rotation of the ipsilateral (left) eye about its 135° axis, frequently
accompanied by a smaller CW rotation of the contralateral {right) eye
about its 45° axis. This class is similar to the rotatory class of Dufosse et
al. (1977) and Nagao et al. (1985), except that the contralateral eye
movement was previously found to be either larger than the ipsilateral eye
movement (Nagao et aL, 1985) or to occur without an accompanying
ipsilateral eye movement (Dufosse et al., 1977). In those studies the
rotatory class movements of the contralateral eye were considered to
result from the inhibitory influence of the flocculus on the pathways from
the ipsilateral anterior canal to the contralateral oblique muscles. No
explanation, however. was offered for the ipsilateral eye movement about
the 135° axis. The present study indicates that the short latency
ipsilateral "rotatory" eye movements are essentially the result of floccular
inhibition of the anterior canal pathways to the ipsilateral vertical recti.
A short latency "vertical" class of eye movements was not found in
the present study and such a class should not, in fact, be expected. The
presence of a "vertical" class of eye movements {rotation about a
nasal-occipital axis) indicates that the flocculus stimulation influenced
both anterior and posterior canal pathways, since rotations about a
nasal-occipital (roll) axis requires the action of both the vertical recti and
the obliques of each eye. Because of the geometrical arrangement of the
vertical recti and oblique muscles. it is not possible to produce a ''vertical''
eye movement by inhibition of only the pathways from the ipsilateral
anterior canal.
152
The contralateral "vertical" eye movements found in the present
study as nearly equal, but opposite rotations of the contralateral eye about
its 45° and 135° axes are apparently the same as the contralateral eye
movements in the vertical class of Dufosse et al. (1977) and Nagao et al.
(1985), but, as shown here, the latency of the contralateral 135' axis
component was significantly longer than that of the contralateral 45° axis
component. This difference implies that the effect on the posterior canal
pathways was less direct and possibly not the result of a direct stimulation
of Purkinje cells. Also, in the present study, the ipsilateral eye movement
accompanying the "vertical" contralateral eye movement was not
"vertical", but consisted predominantly of a larger CCW rotation about the
135° axis. While a combination of short and longer latency responses of
the contralateral eye can account for previous observations of contralateral
"vertical" eye movements, such an explanation for the ipsilateral "vertical"
eye movements reported previously is not available from our observations.
4.4.2 Latency considerations.
The range of average latencies for the short latency classes of eye
movements (15-49 msec) are comparable to those reported in other
studies of eye movements evoked by low intensity stimulation of the
flocculus (39 msec in cat, Sato et al., 1984; 24 msec in monkey, Belknap
and Noda, 1987; 30-40 msec in one rabbit, Dufosse et al.. 1977). These
values are larger than those derivable from electrophysiological studies on
the influence of flocculus stimulation on VOR pathways, but the difference
can be reasonably ascribed to the requirement for temporal summation of
the inhibitory influence of Purkinje cells before the behavioral
consequence of low intensity flocculus stimulation is manifest. In previous
studies in the rabbit, the methods of eye movement measurement
(photographs and 1V tracking systems) precluded obtaining accurate
latency measures. Therefore, some of the discrepancies between the
reported "vertical" eye movements and the directions of eye movements
expected from the actions of the vertical recti and oblique muscles could
be due to a confounding of responses with differing latencies.
!53
Whereas the shorter latency responses can be explained in terms of
preVious findings on flocculo-vestibulo-ocular connectivity, the basis of the
longer latency responses is not known. They could originate from axon
reflex actiVity of antidromically activated afferents or they could arise
from influences of Purkinje cells on vestibular circuits of a higher order
than the three neuron arc canal-ocular paths. Of the several longer
latency eye movements. the most intriguing is the ipsilateral adduction.
which in combination with the accompanying contralateral adduction,
results in a vergence movement. It seems odd that the evoked ipsilateral
adduction should arise from stimulation of floccular compartments FC1
and FC3. which are associated with the ipsilateral vertical recti and
contralateral obliques and not from compartment FC 2. which is
associated with eye movements about the vertical axes. The finding of
long latency eye movements that differ from those predicted from studies
conducted in anaesthetized animals is not surprising since the absence of
anaesthesia permits actiVity to traverse more complicated polysynaptic
pathways. Some of these pathways may not reflect the normal operation of
the nervous system because of the artifice of electrical stimulation, while
others may be normally used, but were not revealed in the anaesthetized
animal. Only future studies will reveal which of the long latency responses
reflect normal signal processing.
4.4.3 Topographical organization of the jlocculus.
The anatomical organization of the floccular white matter into
compartments separated by raphes. as revealed by AChE histochemistry.
is directly related to the physiologically distinguishable classes of eye
movements and probably to specific VOR pathways. The existence of a
zonation of the rabbit flocculus has been proposed by Ito and colleagues
(Dufosse et al .. l977; Ito et al .. 1982b: Nagao et al .. 1985 ) on the basis of
the distribution of sites where microstimulation evoked either different
patterns of eye movements or influenced specific VOR pathways.
Throughout the years, however, the localization and the extent of these
different areas has shown a considerable variability. In addition. the
154
proposed organization of the zonation was contrary to the basic principle
of cerebellar zonation because the "rotatory" zone ran sharply across the
"horizontal" and "vertical" zones (Ito et al., 1982b; Nagao et al., 1985).
The present anatomical findings show that the floccular white
matter is subdivided into five compartments, the three largest of which
can be related to particular canal-ocular pathways on the basis of the short
latency evoked eye movements and the relation of the underlying eye
muscles to particular canals. Since the largest eye movement evoked from
stimulation in FC2 is abduction of the ipsilateral eye, it can be concluded
that this compartment is coupled to the horizontal canal pathway linking
the horizontal canal with the horizontal recti muscles. Similarly, floccular
compartments FC1 and FC3 are coupled to the ipsilateral anterior canal
pathways since the short latency components of the evoked eye
movements are largely the consequence of a combination of an increase in
the activity of the ipsilateral inferior rectus and the contralateral superior
oblique muscles and a decrease in the activity of the ipsilateral superior
rectus and the contralateral inferior oblique muscles. The relations
between the three largest floccular white matter compartments and the
eye movements evoked at short latency are summarized in fig. 4.11 and
4.12.
4.4.4 Comparison of eye nwvement and climbing fiber response axes.
The present microstimulation experiments show that the short
latency eye rotations occur about axes that closely correspond to the best
response axes of the ipsilateral horizontal and anterior semicircular
canals and to the rotation axes of the ipsilateral horizontal and vertical
recti and the contralateral oblique muscles (Ezure and Graf, 1984a.b; Graf
et al, 1988). Moreover, the spatial organization of these eye rotation axes
is also similar to that of the preferred axes characterizing the optic flow
signals reaching the flocculus through the CFs from the dorsal cap and
ventrolateral outgrowth of the inferior olive (Simpson et al., 1981; Graf et
al., 1988; Kusunoki et al., 1990; Kana et al., 1990). One class of CFs is
most strongly modulated by rotational optic flow about the vertical axis
155
156
and is activated predominantly from the ipsilateral eye. The two other
classes of CFs have their axis of best modulation close to the horizontal
plane at an azimuthal angle of either about 135° to the ipsilateral
(dominant) eye, or about 45° to the contralateral (dominant) eye.
Consequently, the short latency abduction movement of the ipsilateral eye
evoked by stimulation in FC2 would result in a strong activation of the
vertical axis class of CFs if the rabbit viewed a stationary world. whereas
the short latency eye movements evoked by stimulation in FC1 and FC3
would result in a strong activation of the ipsilateral 135° axis and the
contralateral 45° axis CF classes.
In this study no distinction was evident between FC1 and FC3 with
regard to the evoked eye movement patterns. For both compartments
electrical stimulation resulted in eye movements whose largest short
latency component was a CCW rotation of the ipsilateral (left) eye about its
135° axis. Also. the range of amplitudes of the accompanying CW 45° axis
component of the contralateral (right) eye was similar for the two
compartments. Although the absence of differences between FC1 and FC3
indicates that they both receive input from both the ipsilateral 135° and
the contralateral 45° classes of CFs. the possibility of a finer organizational
structure in FC1 and FC3 exists because these two classes of CFs have
their origin in different parts of the rostral dorsal cap and ventrolateral
fig. 4.11 Summary diagram of the topographical organization of the three largest compartments delineated by AChE staining and the classes of short latency eye movements evoked by electrical stimulation in the left flocculus. Stimulation of the middle compartment (compartment FC2. medium dots) evoked abduction of the ipsilateral eye with a varying amount of adduction of the contralateral eye. Stimulation of the two compartments on either side of the middle compartment {FC 1, large dots, and FC3, small dots} produced ipsilateral eye movements consisting of a combination of depression, intorsion and abduction, and contralateral eye movements consisting of a combination of elevation, intorsion and adduction. With stimulation of compartments FC 1 and FC3 the movements of the ipsilateral eye are those produced by the vertical recti (inferior rectus contracting), whereas the movements of the contralateral eye are those produced by the oblique muscles (superior oblique contracting}.
157
outgrowth of the inferior olive. In fact, Gerrits and Voogd (1982) have
concluded that in the cat the rostral dorsal cap and the ventrolateral
outgrowth project differentially within the homologous compartments of
FC1 and FC3.
In contrast to the projections of the different climbing fiber classes,
which match the anatomically delineated compartmental organization.
the mossy fiber projections appear to be far less compartmentalized
(Yamamoto, l979a; Sato et al., l983a,b). In addition. one would expect
the mossy fiber input to be distributed across compartments by parallel
fibers whose length is sufficient to traverse 3-4 compartments (Brand et
al., 1976: Mugnaini, 1983). Therefore, it is surprising that the best
response axes for visually induced climbing fiber and simple spike
modulation of floccular Purk.inje cells are often found to be closely aligned
(Graf et al., 1988: Kano et al., 1991). This observation reveals that despite
the absence of an anatomical compartmentation, the mossy fiber input. as
reflected in the simple spike activity of Purk.inje cells, can be functionally
compartmentalized.
The existence of only a small number of classes of both climbing
fiber afferents and evoked eye rotations and their correspondence with
the anatomically distinguishable compartments suggests that the
compartments are the structural correlate of a coordinate system in
which eye movements are represented. Any particular eye rotation can be
synthesized from components taken in this coordinate system, but it
remains to be determined how these components are computed from the
diffusely organized mossy fiber inputs.
4.4.5 Comparison with eye movements evoked by floccular stimulation in cat
In the present study on the flocculus of the awake rabbit, the
classes of short latency evoked eye movements and the distribution of the
respective stimulation sites are in close agreement with the results
obtained by Sato and Kawasaki (l990a,b;199!) With microstimulation in
the flocculus of the anesthetized cat. In their study, division of the
flocculus into three zones (rostral, middle and caudal) emerged from the
!58
distribution of stimulation sites that elicited different classes of eye
movements. In both species the zones or compartments form oblique
strips running, when viewed from above, from caudomedial to
rostrolateral. Electrical stimulation of the middle zone in the cat flocculus
evoked abduction of the ipsilateral eye and adduction of the contralateral
eye, as in the rabbit with stimulation of FCz. Stimulation of the caudal and
rostral zones of the cat flocculus produced a downward movement of both
eyes in combination with an intorsion of the contralateral eye and a small
extorsion of the ipsilateral eye. At first glance, these movements in the
cat appear to differ from those produced by stimulation of FC 1 and FC3 in
the rabbit, but because of differences in the eye ball referenced
kinematics of eye muscles in frontal- and lateral-eyed animals, they, in
fact, result from the action of the same muscles (Simpson and Graf, 1981;
Graf and Simpson, 1981). In both species the movement of the ipsilateral
eye is due to the action of the vertical recti, with the inferior rectus
contracting, whereas the movement of the contralateral eye is due to the
action of the obliques, With the superior oblique contracting. These
similarities between cat and rabbit suggest that in both lateral-eyed and
frontal-eyed species the preferred axes of the visual CF input and the
floccular Purkinje cell output are similarly organized in a coordinate
system whose spatial orientation bears a close resemblance to that of the
semicircular canals and the extra-ocular muscles.
As in the cat, electrical stimulation of the rabbit flocculus delineated
three compartments on the basis of eye movement patterns. While the
AChE staining in rabbit shows that two additional compartments are
present, tbeir relatively small size precluded localization of tbe stimulus
to tbem. A similar limitation was likely present for stimulation of the cat
flocculus because tbe anatomical study by Gerrits and Voogd (1982) of the
cat flocculus showed CF projection zones that are comparable not only to
the three compartments stimulated in the rabbit, but also to the two
smaller compartments, Cz and FC4. On the basis of the source (caudal
dorsal cap, see chapter 2) of the CF projection to FC4 in the rabbit and its
corresponding zone in the cat (Fl), it is likely that this zone is related to
159
horizontal eye movements. The C2 compartment in rabbit (see chapter 2)
and its corresponding zone in the cat (C2) receive from the rostral part of
the medial accessory olive. This part of the flocculus may not be directly related to eye movement control (Ito et al., 1982a,b). A recent anatomical
study (Ruigrok et al., 1992) has shown that the compartmental pattern of
the CF projection to the rat flocculus is like that of rabbit and cat, which
increases the likelihood that the geometrical relation between floccular
compartments, their CF inputs, and particular eye muscles described
above is fundamental to compensatory eye movement control.
fig. 4.12 Summary diagram of the classes of eye movements elicited from the various zones (FZ) in the flocculus.
caudal
160
5. General conclusions
This study demonstrates a reproducible compartmentation in the
white matter of the rabbit flocculus and nodulus with AChE
histochemistry (chapters l and 2). Compact sheets of thin myelinated
AChE-positive fibers. the so-called raphes. separate compartments of
coarser fibers staining less intensely for AChE. ln the flocculus the white
matter was subdiVided in 5 compartments. indicated as Cz, FC1-FC4 from
laterally to medially. The nodulus consisted of 4 compartments, XC1-XC4.
The organization of raphes and compartments was then used to study the
interrelations of afferent climbing fibers and efferent Purkinje cell axons
from different subsets of olivary neurons and Purkinje cells. In chapters 2
and 3 the results of combined anterograde and retrograde tracing
experiments with HRP, WGA-HRP, tritiated leucine. and AChE
histochemistry are reported. They indicate that Purkinje cell populations
in the flocculus and nodulus are portioned into zones that receive their
afferent climbing fiber input from certain subnuclei of the inferior olive
and send efferent projections to specific vestibular and cerebellar
neurons (see diagrams figs. 2.19 and 3.17). The Cz zone projects to the
posterior interposed nucleus and receives climbing fibers from the rostral
pole of the medial accessory olive. The FZI and FZm zones project to the
superior vestibular nucleus and receive their climbing fiber input from
the rostral dorsal cap (rdc) and the ventrolateral outgrowth (vlo). The FZ1
and FZm zones alternate with another pair of zones. These zones FZn and
FZIV project to the medial vestibular nucleus and receive climbing fibers
from the caudal dorsal cap (cdc). Physiological experiments using
microstimulation of the Purkinje cell axons contained in the
compartments showed that two main classes of eye movements can be
linked to two sets of the white matter compartments.
Thus two main functional modules can be distinguished in the
flocculus. each with its own neuronal circuit. Each consists of two
Purkinje cell zones. The VA module controls eye movements around a
vertical axis by the lateral and medial rectus muscles. It consists of the
161
FZn and most likely also the FZIV zones and their projections to
vestibula-ocular relay cells of the horizontal semicircular canal in the
medial vestibular nucleus. It receives climbing fibers from the caudal
dorsal cap, which convey optokinetic information about movements of the
surround around a vertical axis. The 135° axis module consists of the FZI
and FZm zones and their projections to vestibula-ocular relay cells of the
anterior semicircular canal, located in the superior vestibular nucleus
which controls movement of the eyes around a horizontal, 135° axis via
the ipsilateral vertical recti muscles and the contralateral oblique
muscles. It receives a climbing fiber projection from the rostral dorsal
cap and the ventrolateral outgrowth. These climbing fibers carry
optokinetic information about rotations of the surround around a 135°
horizontal axis.
The functional significance of floccular zones C2 and FZJV remains
unclarified because the relatively small size of compartments C2 and FC4
made it difficult to restrict the stimulation to them.
162
6. Abbreuiations
A ANS ANT J3 be bp Cl Cll cc cdc CGM co DAO DC.dc dl DLP dlp dmcc DTN DV F FC FZ m fipl fm, m fp. p fl f2 f3 f4 f5 FLO g !A IP !TN 1 1 VII l!X !X L 16 Lpc LTN LV m MAO ML
=zone A = ansiform lobule = anterior lobule = nucleus beta = brachium conjunctivum = brachium pontis =crus I =crus n = commissura cerebelli = caudal dorsal cap = medial geniculate body = cochlear nucleus = dorsal accessory olive =dorsal cap = dorsal lamella = dorsolateral protuberance =lateral A zone of Buisseret-Delmas (1988) = dorsomedial cell column = dorsal terminal nucleus = descending vestibular nucleus = fastigial nucleus = floccular compartment = floccular zone = primary fissure = posterolateral fissure = folium m of the flocculus =folium p .. = folium l .. = folium 2 ,, = folium 3 , = folium 4 ,, = folium 5 , = flocculus = genu of the facial nerve = anterior interposed nucleus = posterior interposed nucleus :::: interstitial terminal nucleus = lateral = lobule VII = lobule IX. uvula = lobule X, nodulus = lateral cerebellar nucleus = L6wy's fiber bundle = parvicellular part of the lateral cerebellar nucleus = lateral terminal nucleus = lateral vestibular nucleus =folium m = medial accessory olive = medial lemniscus
163
mlf MTN MV MVmc no p PAG PC pf PFL PFLd PFLv PH PMD PO r rb rdc RN r!V sad Sl SN sv t IV vl vlo VTRZ u XC xz y Ill v VI VII Vlll
= medial longitudinal fascicle = medial terminal nucleus = medial vestibular nucleus = magnocellular medial vestibular nucleus = nucleo-olivary tract =folium p = periaqueductal grey = cerebral peduncle = floccular peduncle = paraflocculus = dorsal paraflocculus =ventral paraflocculus = nucleus prepositus hypoglossi = paramedian lobule = principal olive =nucleus r. = restiform body = rostral dorsal cap = red nucleus = lateral recess of the fourth ventricle =dorsal acoustic stria = simplex lobule = substantia nigra = superior vestibular nucleus = taenia of fourth ventricle = ventral lamella = ventrolateral outgrowth = Visual tegmental relay zone = uncinate fascicle = compartment of the nodulus = zone of the nodulus =group y = oculomotor nucleus = trigeminal complex = abducens nucleus = facial nerve = vestibulocochlear nenre
164
7. References
Akagi, Y. (1978) The localization of the motor neurons innervating the extraocular muscles in the oculomotor nuclei of the cat and rabbits, using horseradish peroxidase. J. Comp. Neurol. 181:745-762
Akaike, T. (1986a) Electrophysiological analysis of the tecto-olivocerebellar {crus II) projection in the rat. Brain Res. 378:186-190
Akaike, T. (l986b) Differential localization of Inferior olivary neurons projecting to the tecto-olive-recipient zones of lobule VII or crus I I in the rat cerebellum. Brain Res. 386:400-404
Akaike, T. (1987) Electrophysiological analysis of the tecto-olivocerebellar {lobulus simplex) projection in the rat. Brain Res., 417:371-376
Akaike, T. {1992) The tectorecipient zone in the inferior olivary nucleus in the rat. J. Comp. Neurol. 320:398-414
Alley, K.A. (1977) Anatomical basis for interaction between cerebellar flocculus and brainstem. In R. Baker and A. Berthoz (eds): Control of Gaze by Brain Stem Neurons. Amsterdam: Elsevier, pp.l09-ll7
Alley, K.A., R. Baker, and J.I. Simpson (1975) Afferents to the vestibula-cerebellum and the origin of the Visual climbing fibers in the rabbit. Brain Res. 98:582-589
Altman, J., and G.D. Das (1970) Postnatal changes in the concentration and distrtbution of cholinesterase in the cerebellar cortex of rats. Exp. Neurol. 28: 11-34
Andersson, G., and 0. Oscarsson (1978a) Projections to lateral vestibular nucleus from cerebellar climbing fiber zones. Exp. Brain Res. 32:549-564
Andersson, G., and 0. Oscarsson (1978b) Climbing fiber microzones in the cerebellar vermis and their projection to different groups of cells in the lateral vestibular nucleus. Exp. Brain Res. 565-579
Angaut, P., and A. Broda! (1967) The projection of the "vestibulocerebellum" onto the vestibular nuclei in the cat. Arch. !tal. Biol. 105:441-479
Appleyard, M., and H. Jahnson (1992) Actions of acetylcholinesterase in the guinea pig cerebellar cortex in Vitro. Neuroscience, 47:291-301
Apps. R. (1990) Columnar organization of the inferior olive projection to the posterior lobe of the rat cerebellum. J. Comp. Neurol. 302:236-254
Baker, R., Precht, W., and R. Llinas (1972) Cerebellar modulatory action on the vestibulotrochlear pathway in the cat. Exp. Brain Res. 15:364-385
Baker, R., and R.F. Spencer (1981) Synthesis of horizontal conjugate eye movement signals in the abducens nucleus.
165
Jap. J. EEG. EMG 31:49-59 Baker, R.. N. Mana. and H. Shimazu (1969) Postsynaptic potentials in
abducens motoneurons induced by vestibular stimulation. Brain Res. 15:577-580
Balaban, C.D. (1984} Olive-vestibular and cerebella-vestibular connections in albino rabbits. Neuroscience 12. 129-149
Balaban. C.D., and R.T. Henry (1988) Zonal organization of olive-nodulus projections in albino rabbits. Neurosci. Res. 5:409-425
Balaban, C.D., and E. Watanabe {1984) Functional representation of eye movements in the flocculus of monkeys (Macaca fuscata). Neurosci. Lett. 49:199-205
Balaban. C.D., M. Ito. and E. Watanabe (1981) Demonstration of zonal projections from the cerebellar flocculus to vestibular nuclei in monkeys (Macaca fuscata). Neurosci. Lett 27:101-105
Barmack. N.H., R.W. Baughman. and F.P. Eckenstein (1992a) Cholinergic innervation of the cerebellum of rat. cat, and monkey as revealed by choline acetyltransferase activity and immunohistochemistry. J. Camp. Neural. 317:233-249
Barmack. N.H., R.W. Baughman. F.P. Eckenstein. and H. Sjojaku (1992b) Secondary vestibular cholinergic projection to the cerebellum of rabbit and rat as revealed by choline acetyltransferase immunohistochemistry. J. Comp. Neural. 317:250-270
Barmack. N.H .. M. Fagerson. and P. Errico {1993) Cholinergic projection to the dorsal cap of the inferior olive of the rat. rabbit, and monkey. J. Camp. Neural. 328:263-281
Barth. F.. and M.S. Ghandour (1983) Cellular localization of butyrylcholinesterase in adult rat cerebellum determined by immunofluorescence. Neurosci. Lett. 39: !49-153
Belknap. D.B., and H. Noda (!987) Eye movements evoked by microstimulation in the flocculus of the alert macaque. Exp. Brain Res. 67:352-362
Bernard. J.F. (1987) Topographical localization of olivocerebellar and corticonuclear connections in the rat - a WGA-HRP study. I. Lobules IX, X. and the flocculus. J. Camp. Neurol. 263:241-258
Bigare, F. {1980). De efferente verbindingen van de cerebellaire schors van de kat. Thesis
Blanks, R.H.I. (1990) Afferents to the cerebellar flocculus in cat with special reference to pathways conveying vestibular. visual (optokinetic) and oculomotor signals. J. Neurocytol. 19:628-642
Blanks. R.H.I.. W. Precht. and Y Torigoe (1983) Afferent projections to the cerebellar flocculus in the pigmented rat demonstrated by retrograde transport of horseradish peroxidase. Exp. Brain Res. 52:293-306
166
Boegman. R.J .. A. Parent. and R. Hawkes (1988) Zonation in the rat cerebellar cortex: patches of high acetylcholinesterase activity in the granular layer are congruent with Purkinje cell compartments. Brain Res. 448:237-251
Balk, L. (1906). Das Cerebellum der Saugetiere: Eine vergleichende anatomische Untersuchung. Jena/Haarlem, Fischer.
Brand. S. A..-L. Dahl. and E. Mugnaini (1976) The length of parallel fibers in the cat cerebellar cortex. An experimental light and electron microscopic study. Exp. Brain Res. 26, 39-58.
Broda!, A. (1940) The cerebellum of the rabbit. A topographical atlas of the folia as revealed in transverse sections. J. Camp. Neural. 72:63-81
Broda!. A. (1976) The olivocerebellar projection in the cat as studied with the method of retrograde axonal transport of horseradish peroxidase. II. The projection to the uvula. J. Camp. Neural. 166:417-426
Broda!. P., and A. Broda! (1981) The olivocerebellar projection in the monkey. Experimental studies with the method of retrograde tracing of horseradish peroxidase. J. Camp. Neural. 201:375-393
Broda!, P., and A. Broda! (1982) Further observations on the olivocerebellar projections in the monkey. Exp. Brain Res. 45:71-83
Brodal, A., and K. Kawamura (1980) Olivocerebellar projection: a review. Adv. Anal. Embryo!. Cell Biol. 64:1-137
Broda!, A.. and 0. Pompelano (1957) The vestibular nuclei in the cat. J. Anal. 91:438-454
Brown, W.J .. and S.L. Palay (1972) Acetylcholinesterase activity ln certain glomeruli and Golgi cells of the granular layer of the rat cerebellar cortex. Z. Anal. Entwlckl.-Gesch. 137:317-334
Buisseret-Delmas, C. (1988) Sagittal organization of the olivocerebellonuclear pathway in the rat. I. Connections with the nucleus fastigii and the nucleus vestibularis. Neurosci. Res. 5:475:493
Cajal, S. Ramon y (1911) Histologie du systeme nerveux de l'homme et de vertebres. Vol. II. Parts: Maloine, p. 24
Carleton, S.C., and M.B. Carpenter (1984) Distribution of primary vestibular fibers in the brain stem and cerebellum of the monkey. Brain Res. 294:381-298
Carpenter, M.B., and R.J. Cowie (1985) Connections and oculomotor projections of the superior vestibular nucleus and cell group 'y'. Brain Res. 336:265-287
Chambers, W.W., and J.M. Sprague (1955a) Functional localization in the cerebellum. I. Organization in longitudinal cortico-nuclear zones and their contribution to the control of posture. both extrapyramidal and pyramidal. J. Camp. Neural. 103: 105-130
Chambers, W.W., and J.M. Sprague (1955b) Functional localization in the cerebellum. II. Somatotopic organization in cortex and nuclei.
167
Arch. Neurol. Psychiat. 74:653-680 Chan-Palay. V .. G. Nilaver. S.L. Palay. M.C. Beinfeld. E.A. Zimmerman. J.Y.
Wu, and T.L. O'Donohue (1981) Chemical heterogeneity in cerebellar Purkinje cells: Existence and coexistence of glutamic acid decarboxylase-like and motilin-Uke immunoreactiVities. Proc.Natl.Acad.Sci.USA 78:7787-7791
Chan-Palay. V .. S.L. Palay. and J.Y. Wu (1982) Sagittal cerebellar microbands of taurine neurons: Immunocytochemical demonstration by using antibodies against the taurine synthetizing enzyme cysteine sulfinic acid decarboxylase. Proc.NatlAcad.Sci.USA 79:4221-4225
Chubb. M.C .. and A.F. Fuchs (1982) Contribution of y group of vestibular nuclei and dentate nucleus of cerebellum to generation of vertical smooth eye movements. J. Neurophysiol. 48:75-99
Chubb. M.C .. A.F. Fuchs. and C.A. Scudder (1984) Neuron actiVity in monkey vestibular nuclei during vertical vertical vestibular and eye movements. J. Neurophysiol. 52:724-742
Collewijn. H (1977) Eye and head movements in freely moVing rabbits. J. Physiol 266:471-498
Courville. J .. F. Faraco-Cantin. and N. D!akiw (1974) A functionally important feature of distribution of the olivocerebllar climbing fibers. Can. J. Physiol. Pharmacal. 52:1212-1217
Csill!k. B .• F. Joo and P. Kasa (1963) Cholinesterase activity of archicerebellar mossy fiber apparatus. J. Histochem. Cytochem. 11. 103-!13.
Del..acalle. S .• LB. Hersch, and C.B. Saper (1993) Cholinergic innervation of the human cerebellum. J. Comp. Neural. 328:364-376
De Zeeuw. C.!.. P.L. DiGiorgi. D.R. Wylie. D. Wang. E. Marsh. and J.l. Simpson (1992) Projections of indiVidual Purkinje cells of identified zones in the rabbit flocculus to the vestibular and cerebellar nuclei. Soc. Neurosci. Abstr. 18:854
De Zeeuw. C.!., P. Wentzel, and E. Mugnaini (1993) Fine structure of the dorsal cap of the inferior olive and its GABAergic and non-GABAergic input from the nucleus prepositus hypoglossi in rat and rabbit. J. Camp. Neural. 327:63-82
Dow, R.S. (1936) The fiber connections of the posterior parts of the cerebellum in the cat and rat. J. Comp. Neural. 63:527-548
Dow, R.S. (1938a) Efferent connections of the flocculonodular lobe in Macaca mulatta. J. Comp. Neural. 68:297-305
Dow. R.S (l938b) Effects of lesions in the vestibular part of the cerebellum in primates. Arch. Neurol. Psychiat. 40:500-520
Dow, R.S. (l-939) Cerebellar action potentials in response to stimulation of various afferent connections.
168
J. Neurophysiol. 2:543-555 Dufosse. M .. M. Ito, andY. Miyashita (1977) Functional localization in the
rabbit's cerebellar flocculus determined in relationship with eye movements. Neurosci. Lett. 5:273-277
Eisenman, L.M. (1984) Organization of the olivocerebellar projection to the uvula in the rat. Brain Behav. Evol. 24:1-12
Epema, A. H. (1990) Connections of the vestibular nuclei in the rabbit. Thesis, Rotterdam
Epema, A.H., N.M. Gerrits, and J. Voogd (1988) Commissural and intrinsic connections of the vestibular nuclei in the rabbit: a retrograde tracer study. Exp. Brain Res. 71:129-146
Epema, A.H., N.M. Gerrits, and J. Voogd (1990) Secondary vestibulocerebellar mossy fiber projections to the flocculus and uvulonodular lobule in the rabbit: a study using HRP and double fluorescent tracer techniques. Exp. Brain Res. 80:72-82
Friede, R.L., and L.M. Fleming (1964) A comparison of cholinesterase distribution in the cerebellum of several species. J. Neurochem. 11:1-7
Fukuda, J., S.M. Highstein, and M. Ito (1972) Cerebellar inhibitory control of the vestibula-ocular reflex investigated in rabbit Hlrd nucleus. Exp. Brain Res. 14:511-526
Furber, S.E., and C.R.R. Watson (1983) Organization of the olivocerebellar projection in the rat. Brain Behav. Evol. 22:132-152
Gacek, R.R. (1977) Location of brain stem neurons projecting to the oculomotor nucleus in the cat. Exp. Neural. 57:725-749
Gacek, R.R. (1979) Location of abducens afferent neurons in the cat. Exp. Neural. 64:342-353
Geneser-Jensen, F.A., and T.W. Blackstad (1971) Distribution of acetylcholinesterase in the hippocampal region of the guinea-pig. I. Entorhinal area, parasubiculum and presubiculum. Z. Zellforsch. Mikrosk. Anat. 114:460-481
Gerebtzoff, M.A. (1959) Cholinesterases: A histochemical contribution to the solution of some functional problems. Internal. Series of Monographs on Pure and Applied Biology, Vol. 3. Oxford: Pergamom Press.
Gerrits, N.M., and J. Voogd (1982) The climbing fiber projection to the flocculus and adjacent paraflocculus in th cat. Neuroscience 7: 2971-2991
Gerrits, N.M., and J. Voogd (1986) The nucleus reticularis tegmenti pontis and the adjacent paramedian pontine reticular formation: differential projections to the cerebellum and the caudal brainstem. Exp. Brain Res. 62:29-45
Gerrits, N.M., and J. Voogd (1989) The topographical of climbing and mossy fiber afferents in the flocculus and the ventral paraflocculus
169
in rabbit, cat and monkey. In: P. Strata (ed): The Olivocerebellar System in Motor Control. Exp. Brain Res. Series Vol. 17. Amsterdam: Elsevier. pp. 26-29
Gerrits, N.M., A.H. Epema, A. van Linge, and E. Dalm (1989) The primary vestibulocerebellar projection in the rabbit: absence of primary afferents in the flocculus. Neurosci. Lett. 290:262-277
Gerrits, N.M .. J. Voogd, and l.N. Magras (1985) Vestibular afferents of the inferior olive and the vestibula-cerebellar climbing fiber pathway to the flocculus in the cat. Brain Res. 332:325-336
Gibson, A.R .. D.!. Hansma, J.C. Houk, and F.R. Robinson (1984) A sensitive low artifact TMB procedure for the demonstration of WGA-HRP in the CNS. Brain Res. 298:235-241
Giolli. R.A., R.H.l. Blanks. and Y. Torigoe ( 1984) Pretectal and brain stem projections of the medial terminal nucleus of the accessory optic system of the rabbit and rat as studied by anterograde and retrograde neuronal tracing methods. J. Camp. Neural. 227:228-251
Giolli, R.A., R.H.I. Blanks. Y. Torigoe. and D.D. Williams (1985) Projections of medial accessory optic nucleus, ventral tegmental nuclei. and substantia nigra of rabbit and rat as studied by retrograde axonal transport of horseradish peroxidase. J. Camp. Neural. 232:99-116
Giolli. R.A., Y. Torigoe, R.H.l. Blanks, and H.M. McDonald (1988) Projections of the dorsal and lateral terminal accessory optic nuclei and of the interstial nucleus of the superior fascicle (posterior fibers) in the rabbit and rat. J. Camp. Neural. 277:608-620
Goodman. D.C., R.E. Hallett. and R.B. Welch (1963) Patterns of localization in the cerebellar cortico-nuclear projection of the albino rat. J. Comp. Neural. 121:51-67
Gorenstein, C., N.C. Bundman. J.L. Bruce. and A. Rotter (1987) Neuronal localization of pseudocholinesterase in the rat cerebellum: saggital bands of Purkinje cells in the nodulus and uvula. Brain Res. 418:68-75
Graf. W., and J.l. Simpson (1981) The relations between the semicircular canals, optic axis, and the extraocular muscles in lateral-eyed and frontal-eyed animals. In A. Fuchs and W. Becker (eds): Progress in Oculomotor Research. Amsterdam: Elsevier, pp. 411-420
Graf. W., J.l. Simpson. and C.S. Leonard (1988) Spatial organization of visual messages of the rabbit's cerebellar flocculus. II. Complex and simple spike responses of Purkinje cells. J. Neurophysiol. 60:2091-2121
Graham. R.C., and M.J. Kamovsky (1966) The early stages of absorption of injected horseradish peroxidase in the proximal tubules of the mouse kidney. Ultrastructural cytochemistry by anew technique. J. Histochem. Cytochem.14:29l-302
170
Graybiel. A.M .. and E.A. Hartwieg (1974) Some afferent connections of the oculomotor complex in the cat: an experimental study with tracer techniques. Brain Res. 81:543-551
Greenfield, S.A (1984) Acetylcholinesterase may have novel functions in the brain. Trends in Neurosci. 7:364-368
Groenewegen. H.J .. and J. Voogd (1977) The parasagittal zonation within the olivocerebellar projection. l. Climbing fiber distribution in the vermis of the cat cerebellum. J. Camp. Neural. 174:417-488
Groenewegen, H.J., J. Voogd, and S.L. Freedman (1979) The parasagittal zonation within the olivocerebellar projection. II.Climbing fiber distribution in the intermediate and hemispheric parts of the cat cerebellum. J. Camp. Neural. 183: 551-602
Gulya, K., and P. Kasa (1983) Structural localization of the elements of the acetylcholine system in the rat cerebellum. Acta Biol. Hung. 34:155-161
Haggqvist, G. ( 1936) Analyse der Faserverteilung in einem Ruckenmarkquerschnitt (Th. 3). Z. mikr. anat. Forschung 29:1-34
Haines, D.E. (1977) Cerebellar corticonuclear and corticovestibular fibers of the flocculonodular lobe in a prosimian primate (Galago senegalensis). J. Camp. Neural. 174:607-630
Hawkes, R., and N. Leclerc (1987) Antigenic map of the rat cerebellum: distribution of parasagittal bands as revealed by monoclonal anti-Purkinje cell antibody mabQ 113. J. Camp. Neural. 256:29-41
Hess, D.T., and J. Voogd (1986) Chemoarchitectonic zonation of the monkey cerebellum. Brain Res. 369:383-387
Highstein, S.M. (1971) Organization of the inhibitory and excitatory vestibule-ocular reflex pathways to the third and fourth nuclei in rabbit. Brain Res. 32:218-224
Highstein, S.M. (1973) The organization of the vestibula-ocular and trochlear reflex pathways in the rabbit. Exp. Brain Res. 17:285-300
H!ghstein, S.M., and H. Reisine (1979) Synaptic and funtional organization of vestibula-ocular reflex pathways. Prog. Brain Res. 50:431-442
Hoddevlk, G.H. and A. Broda! (1977) The olive-cerebellar projections studied with the method of retrograde axonal transport of horseradish peroxidase. V. The projection to the floccule-nodular lobe and the paraflocculus in the rabbit. J. Camp. Neural. 176: 269-280
Holstege, G., and H. Collew!jn (1982) The efferent connections of the nucleus of the optic tract and the superior colliculus in the rabbit. J. Camp. Neurol. 209:139-175
171
Hunt. S., and J. Schmidt (1978) Some observations on the binding pattern of a.-bungarotoxin in the central nervous system of the rat. Brain Res. 157:213-232
Ikeda, M., T. Houtani, T. Ueyama, and T, Sugimoto (1991) Choline acetyltransferase immunoreactivity in the cat cerebellum. Neuroscience, 45:671-690
!!ling. R.-B. (1990) A subtype of cerebellar Golgi cells may be cholinergic. Brain Res. 522:267-274
Ingram, V.M .. M.P. Ogren, C.L. Chatot, J.M. Gossels, and B.B. Owens (1985) Diversity among Purkinje cells in the monkey cerebellum. Proc. Nat!. Acad. Sci. USA 82:7131-7135
Ito, M. (1972) Neural design of the cerebellar motor control system. Brain Res. 40:81-84
Ito, M. (1982) Cerebellar control of the vestibula-ocular reflex-around the flocculus hypothesis. Ann. Rev. Neurosci. 5:275-296
!to, M. (1984) The cerebellum and neuronal control. Raven, New York. !to, M. (1990) A new physiological concept on cerebellum.
Rev. Neural. 146:564-569 !to, M., S.M. Highstein, and J. Fukuda (1970) Cerebellar inhibition of the
vestibula-ocular reflex in rabbit and cat andits blockage by picrotoxin. Brain Res. 17:524-526
!to, M., P.J. Jastreboff, and Y. Miyashita (1982a) Specific effects of unilateral lesions in the flocculus upon eye movements in albino rabbits. Exp. Brain Res. 45:233-242
Ito, M .. Y. Miyashita, and A. Ueki (1978) Functional localization in the rabbit's inferior olive determined in connection with the vestibula-ocular reflex. Neurosci. Lett. 8:283-287
Ito, M., N. Nisimaru, and M. Yamamoto (1973) Specific neural connections for the cerebellar control of vestibulo-ocularreflexes. Brain Res. 60:238-243
Ito, M., N. Nisimaru. and M. Yamamoto (!977) Specific patterns of neuronal connections involved in the control of the rabbit's vestibula-ocular reflexes by the cerebellar flocculus. J. Phys. (Lond.) 265:833-854
Ito, M., I. Orlov, and M. Yamamoto (l982b) Topographical representation of vestibula-ocular reflexes in rabbit cerebellar flocculus. N eurosci. 7: 1644-165 7
Jansen, J., and A. Brodal (1940) Experimental studies on the intrinsic fibers of the cerebellum. H. The cortico-nuclear projection. J. Camp. Neural. 73:267-321 Jansen, J. and Brodal, A. (1942) Experimental studies on the intrinsic fibers of the cerebellum. III. Cortico-nuclear projection in the rabbit and the monkey. Norske Vid. Akad. Avh.l. Math. Nat.Kl. No.3: 1-30.
Kan, K.S.K., L.P. Chao, and L.S. Forno (1978) Immunohistochemical localization of choline acetyltransferase in rabbit spinal cord and
172
cerebellum. Brain Res. 146:221-229
Kan, K.S.K., L.P. Chao. and L.S. Forno (1980) Immunohistochemical localization of choline acetyltransferase in the humancerebellum. Brain Res. 193:165-171
Karczmar. A.G. (1969) Is the central cholinergic nervous system overexploited ? Fedn. Proc. Fedn. Am. Sacs. Exp. Bioi. 28:147-157
Kanda, K-1.. Y. Sato, K. Ikarash!, and T. Kawasaki (1989) Zonal organization of climbing fiber projections to the uvula in the cat. J. Camp. Neural. 279:138-148
Kana, M-S., M. Kana, and K. Maekawa (1990) Receptive field organization of climbing fiber afferents responding to optokinetic stimulation in the cerebellar nodulus and flocculus of the pigmented rabbit. Exp. Brain Res. 82:499-5!2
Katayama, S., and W. Nisimaru (1988) Parasag!ttal zonal pattern of olivo-nodular projections in rabbit cerebellum. NeuroscL Res. 5:424-438
Kaufman, G.D., J.H. Anderson and A. Beitz (1991) Activation of a specific vestibula-olivary pathway by centripetal acceleration in rat. Brain Res. 562: 311-317.
Kawaguchi, Y (!985) Two groups of secondary vestibular neurons mediating hortzontal canal signals, probably to the ipsilateral medial rectus muscle, under inhibitory influences from the cerebellar flocculus in rabbits. Neurosci. Res. 2:434-446
Kawamura. K .. and T. Hashikawa (1979) Olivocerebellar projections in the cat studied by means of anterograde axonal transport of labeled amino acids as tracers. NeuroscL 4:1615-1633
Kevetter, G.H., and AA Perachio (!984) Central projections of vestibular afferents innervating the macula of the saccule ingerbil. Neurosci. Lett. 5! :7-12
Kimura, H., P.L. McGeer, J.H. Peng, and E. McGeer (!98!) The central cholinergic system studied by choline acetyltransferase immunohistochemistry in the cat. J. Camp. Neural. 200:151-201
Kame!, 1., T. Haja!, P.L. McGeer, and E.G. McGeer (1983) Evidence for an !ntracerbellar acetylcholinesterase-rich but probably non-cholinergic flocculonodular projection in the rat. Brain Res. 258:115-1!9
Kornel!ussen, H.K. ( 1968) On the ontogenic development of the cerebellum (nuclei, fissures, and cortex) of the rat, with special reference to regional variations in corticogenesis. J. Hirnforsch. 10:379-412
Kooy, F.H. (1916) The inferior olive in vertebrates. Thesis.Haarlem: Bohn.
Kuhar, M.J., and H.!. Yamamura (1975) Light autoradiographic localization of cholinergic muscarinic receptors in rat brain by specific binding of a potent antagonist. Nature 253:560-561
173
Kusunoki, M., M. Kano, M.-S. Kano, and K. Maekawa (1990) Nature of optokinetic response and zonal organization of climbing fiber afferents in the vestibulocerebellum of the pigmented rabbit. I. The flocculus. Exp. Brain Res. 80:225-237
Labandeira-Garcia, J.L., M.J. Guerra-Seijas, J.A. Labandeira-Garcia, and F.J. Jorge-Barreira (1991) Distribution of the vestibular neurons projecting to the oculomotor and trochlear nuclei in rabbits. Brain Behav. Evol. 37:111-124
Langer, T. (1985) Basal interstitial nucleus of the cerebellum: cerebellar nucleus related to the flocculus. J. Comp. Neurol. 235:38-47
Langer, T., A.F. Fuchs, C.A. Scudder, and M.C. Chubb (l985a) Afferents to the flocculus of the cerebellum in the Rhesus Macaque as revealed by retrograde transport of horseradish peroXidase. J. Comp. Neurol. 235:1-25
Langer, T .. A.F. Fuchs, M.C. Chubb, C.A. Scudder. and S.G.Lisberger (1985b} Floccular efferents in the Rhesus Macaque as revealed by autoradiography and horseradish peroxidase. J. Comp. Neurol. 235:26-37
Larsell, 0. (1934) Morphogenesis and evolution of the cerebellum. Arch. Neural. Psychiat. (Chic.), 31: 373-395.
Larsen, 0 (1937) The cerebellum, a reView and interpretation. Arch. Neural. Psychiatr. 38:580-607
Larsell, 0 (1952) The morphogenesis and adult pattern of the lobules and fissures of the cerebellum of the white rat. J. Comp. Neural. 97:281-356
Larsell, 0. (1970) The Comparative Anatomy and Histology of the Cerebellum from Monotremes through Apes. J. Jansen (ed.). Minneapolis. University of Minnesota Press.
Larsell, 0., and J. Jansen (1972) The comparative anatomy and histology of the cerebellum. In: The Human Cerebellum, Cerebellar Connections, and The Cerebellar Cortex. Minneapolis. University of Minnesota Press.
Lechan, R.M., J.L. Nestler, and S. Jacobson (1981) Immunohistochemical localization of retrogradely andanterogradely transported wheat germ agglutinin within the central nervous system of the rat: application to immunostaining of a second antigen within the same neuron. J. Histochem. Cytochem. 29:1255-1262
Legendre, A. and J. Courville (1987) Original trajectory of the cerebella-olivary projection: an experimental study with radioactive and fluorescent tracers in the cat. Neurosc!. 21:877-891
Leonard, C.S., J.l. Simpson, and W. Graf (1988) Spatial organization of Visual messages of the rabbit's cerebellar flocculus. I. Typology of inferior olive neurons of the dorsal cap of Kooy. J. Neurophysiol. 60:2073-2090
Lorente de No, R. (1933) Anatomy of the eight nerve. The Laryngoscope, XLIIl: l-38
174
Uiwy, R. (1928) Dber die Faseranatomie und Physiologie der Formatio vermicularis cerebelli. Arb. Neural. lnst. Univ. Wien. 21:359-382
Maekawa. K .. and J .I. Simpson ( 1972) Climbing fiber activation of Purkinje cells in the flocculus by impulses transferred through the visual pathway. Brain Res. 39:245-251
Maekawa. K .. and J.l. Simpson (1973) Climbing fiber responses evoked in the vestibulocerebellum of rabbit from the Visual system. J. Neurophysiol. 36:649-666
Maekawa. K. and T. Takeda (1976) Electrophysiological identification of climbing and mossy fiber pathways from the rabbit's retina to the contralateral cerebellar flocculus. Brain Res. 109:169-174
Maekawa, K. and T. Takeda (1977) Afferent pathways from the visual system to the cerebellar flocculus in the rabbit. ln R.Baker and A. Berthoz (eds): Control by Gaze of Brain Stem Neurons. Amsterdam: Elsevier. pp. 187-195
Maekawa. K., and T. Takeda (1979) Origin of descending afferents to the rostral part of the dorsal cap of inferior olive which transfers contralateral optic activities to the flocculus. A horseradish peroxidase study. Brain Res. 172:393-405
Maekawa. K .. T. Takeda. M. Kana. and M. Kusunokl (1989) Collateralized climbing fiber projection to the flocculus and the nodulus of the rabbit. Exp. Brain Res. Ser. 17:30-45
Magras. LN., and J. Voogd (1985) Distribution of secondary vestibular fibers in the cerebellar cortex. An autoradiographic study in the cat. Acta Anal. (Basel) 123:51-57
Mallo!. J .. M.C. Sarraga, M. Bartolome. M.S. Chandour. and G. Gombos (1984) Muscarinic receptor during postnatal development of rat cerebellum: an index of cholinergic synapse formation ? J. Neurochem. 42:1641-1649
Maran!. E. (1982) Topographic Enzyme Histochemistry of the Mammalian Cerebellum 5'-nucleotidase and Acetylcholinesterase. Thesis, Leiden.
Maran!. E. (1986) Topographic histochemistry of the cerebellum. Prog. Histochem. Cytochem. Vol. 16
Maran!, E .. and J. Voogd (1977a) An acetylcholinesterase bandpattern in the molecular layer of the cat cerebellum. J. Anal. 124:335-345
Maran!. E.. and J. Voogd (!977b) A longitudinal pattern of acetylcholinesterase in the molecular layer of the cerebellum of the cat. Acta Morpho!. Neerl. Scand. 15:332-333
Martin, G.F .. J. Culberson. C. Laxson, M. Linauts, M. Panneton, and I. Tschismadia ( 1980) Afferent connections of the inferior olivary nucleus with preliminary notes on its development: Studies using the North American opossum. In J. Courville, C. C. de Montigny, and Y. Lamarre (eds): The Inferior Olivary Nucleus: Anatomy and
175
Physiology. New York: Raven Press, pp. 35-72 Mash, D.C .. and L.T. Potter (1986) Autoradiographic localization of M I
and M2 muscartne receptors in the rat brain. Neurosci. 19:551-564
McCrea, R.A., and R. Baker (1985) Anatomical connections of the nucleus prepositus hypoglossi of the cat. J. Camp. Neural. 237:377-407
Mesulam, M.M. (1978) Tetramethylbenzidine for horseradish peroxidase neurohistochemistry: a non-carcinogenic blue reaction-product with superior sensitivity for visualizing neural afferents and efferents. J. Histochem. Cytochem. 26:106-117
Mizuno, N., K. Mochizuki, C. Akimoto, and R. Matsushima (1973) Pretectal projection to the inferior olive in the rabbit. Exp. Neural. 39:498-506
Mizuno. N., Y. Nakamura, and N. lwahori (1974) An electronmicroscopic study of the dorsal cap of the inferior olive in the rabbit, with special reference to the pretecto-olivary fibers. Brain Res. 77:385-395
Mugnainl, E. (1983) The length of cerebellar fibers in chicken and rhesus monkey. J. Camp. Neural. 220, 7-15
Nagao, S. (1983) Effects of vestibulocerebellar lesions upon dynamic characteristics and adaptation of vestibula-ocular and optokinetic responses in pigmented rabbits. Exp. Braln Res. 53:36-46
Nagao, S. (1988) Behavior of floccular Purkinje cells correlated with adaptation of horizonatal optokinetic eye movement response in pigmented rabbits. Exp. Braln Res. 73:489-497
Nagao, S., M. Ito, and L. Karachot (1985) Eye field in the cerebellar flocculus of pigmented rabbits determined with local electrical stimulation. Neurosci. Res. 3:39-51
Naito, H., K. Tanimura. N. Taga, and Y. Hosoya (1974) Microelectrode study on the subnuclei of the oculomotor nucleus in the cat. Brain Res. 81:215-231
Nelson, B.J., N.H. Barmack, and E. Mugnaini (1986) GABAergic projection from vestibular nuclei to rat inferior olive. Soc. Neurosci. Abstr. 71:15
Nelson, B.J. and E. Mugnaini (1989) Origins of GABAergic inputs to the inferior olive. In: P. Strata (ed): The Olivocerebellar System in Motor Control. Exp. Brain Res. Series Vol. 17. Amsterdam: Elsevier, pp. 86-107.
Neustadt, A., A. Frostholm, and A. Rotter (1988) Topographical distribution of muscarinic cholinergic receptors in the cerebellar cortex of the mouse, rat, guinea pig, and rabbit: a species comparison. J. Comp. Neurol. 272:317-330
Nilaver. G., R. Defendini, K.A. Zimmerman, M.C. Beinfeld. and T.L. O'Donohue (1982) Motilin in the Purk!nje cell of the cerebellum.
176
Nature 295:597-598 Ojima. H., S.l. Kawajiri, and T. Yamasaki [1989) Cholinergic innervation
of the rat cerebellum: Qualitative and quantitative analyses of elements immunoreactive to a monoclonal antibody against choline acetyltransferase. J. Comp. Neurol. 290:41-52
Ono, M .. and H. Kato (1938) Zur Kenntnis des Kleinhirnkerne des Kaninchens. Anat. Anz. 86:245-259
Oscarsson, 0 [1969) The sagittal organization of the cerebellar anterior lobe as revealed by the projection patterns of the climbing fiber system. In R. Llinas (ed): Neurobiology of Cerebellar Evolution and Development. Chicago, Illinois, AMA, pp. 525-537
Oscarsson. 0. (1973) Functional organization of spinocerebellar paths. In A. Iggo [ed) Handbook of Sensory Physiology. Vol. Ill. Somatosensory System. New York: Springer, pp. 339-380
Oscarsson, 0 (1979) Functional units of the cerebellum-sagittal zones and microzones. Trends Neurosci. 2:143-145
Robertson, L.T., and K. Logan [1986) Relationship of parasagittal bands of acetylcholinesterase actiVity to the climbing fiber representation. Neurosci. Lett. 72:128-134
Robinson, D.A. (1963) A method of measuring eye movement using a scleral search coil in a magnetic field. IEEE Trans. Biomed. Electron. B.E. 10:137-145
Robinson, D.A. [1976) Adaptive gain control of vestibulo-ocular reflex by the cerebellum. J. Neurophysiol. 39:954-969
Ron. S., and D.A. Robinson [1973) Eye movements evoked by cerebellar stimulation in the alert monkey. J. Neurophysiol. 36:1004-1022
Rotter, A., N.J.M. Birdsall, A.S.V. Burgen, P.A. Field, A. Smolen, and G. Raisman (1979a) Muscarinic receptors in the central nervous system of the rat. IV. A comparison of the effects of axotomy and deafferentation on the binding of 3H-propylbenzilcholine mustard and associated synaptic changes in the hypoglossal and pontine nuclei. Brain Res. Rev. 1:207-224
Rotter, A., N.J.M. Birdsall, P.M. Field, and G. Raisman (1979b) Muscarinic receptors in the central nervous system of the rat. II. Distribution of binding of 3H-propylbenzilcholine mustard in the midbrain and hindbrain. Brain Res. Rev. 1:167-183
Ruigrok, T.J.H .. and J. Voogd [1990) Cerebellar nucleo-olivary projections in the rat: an anterograde tracing study With Phaseolus vulgaris-ieucoagglutinin [PHA-L). J. Comp. Neurol. 298:315-333
Ruigrok, T. J. H .. R.J. Osse, and J. Voogd [1992) Organization of inferior oliva:ry projections to the flocculus and ventral paraflocculus of the rat cerebellum. J. Comp. Neurol. 316:129-150
177
Sa!nt-Cyr. J.A., and J. Courville (1979) Projection from the vestibular nuclei to the inferior olive in the cat. An autoradiographic and horseradish peroXidase study. Brain Res. 165:189-200
Sato, Y .. and N.H. Barmack (1985) Zonal organization of olivocerebellar projections to the uvula in rabbits. Brain Res. 359:281-292
Sato. Y .. and T. Kawasaki (1984) Functional localization In the three zones related to eye movement control in the cat. Brain Res. 290:25-31
Sato, T., and T. Kawasaki (1987) Target neurons of floccular caudal zone inhibition in y-group nucleus of vestibular nuclear complex. J. Neurophyslol. 57:460-480
Sato, Y., and T. Kawasaki (1990a) Eye movement evoked by stimulation of Purk.inje cell zones of the cerebellar flocculus in the cat. Acta Med. Bioi. 38:27-35
Sato, Y .. and T. Kawasaki (l990b) Operational unit responsible for plane-specific control of eye movement by cerebellar flocculus in cat. J. Neurophys!ol. 64:551-564
Sato. Y .. and T. Kawasaki (1991) Identification of the Purk!nje cell/climbing fiber zone and its target neurons responsible for eye-movement control by the cerebellar flocculus. Brain Res. Rev. 16: 39-64
Sato, Y., T. Kawasaki, and K. lkarash! (l982a) Zonal organization of the floccular Purkinje cells projecting to the vestibular nucleus in cats. Brain Res. 232:1-15
Sato. Y., T. Kawasaki, and K. lkarash! (1982b) Zonal organization of the floccular Purk!nje cells projecting to group y of the vestibular complex and the lateral cerebellar nucleus in cats. Brain Res. 234:430-434
Sato, Y., T. Kawasaki. and K. Ikarashl (1983a) Afferent projections from the brainstem to the three floccular zones in cats. I. Climbing fiber projections. Brain Res. 272:27-36
Sato. Y., T. Kawasaki, and K. Ikarash! (1983b) Afferent projections from the brainstem to the three floccular zones in cats. 11. Mossy fiber projections. Brain Res. 272:37-48
Sato, Y., F. Yamamoto, H. Shojaka, and T. Kawasaki (1984) Pathway from floccular caudal zone contributing to vertical eye-movements in cats. Role of group y nucleus of vestibular nuclei. Brain Res. 294: 375-380
Sato, Y., K. Kanda. and T. Kawasaki (1988) Target neurons of floccular middle zone inhibition in the medial vestibular nucleus. Brain Res. 446:225-235
Scott. T.G. (1963) A unique pattern of localization In the cerebellum. Nature 200:793
Scott. T.G. (1964) A unique pattern of localization Within the cerebellum of the mouse. J. Comp. Neurol. 122:1-8
178
Silver, A. {1967) Cholinesterases of the central nervous system with special reference to the cerebellum. Int. Rev. Neurobiol. 10:57-109
Silver. A. (1974) The Biology of Cholinesterases. Amsterdam: Elsevier. pp. 355-388
Simpson. J.l.. and K.E. Alley (1974) Visual climbing fiber input to rabbit vestibula-cerebellum: a source of direction-specific information. Brain Res 82:302-308
Simpson. J.I.. W. Graf. and C. Leonard (1981) The coordinate system of visual climbing fibers to the flocculus. In W. Becker and A. Fuchs (eds): Progress in Oculomotor Research. Amsterdam: Elsevier, pp. 475-484
Simpson, J.I., and W. Graf (1985) The selection of reference frames by nature and its investigators. In A. Berthoz and G. Melvill Jones (eds): Adaptive Mechanisms in Gaze Control. Amsterdam, Elsevier, pp. 3-16
Simpson. J.l., W. Graf. and C.S. Leonard (1989) Three-dimensional representation of retinal image movement by climbing fiber activity. In P. Strata (ed): The Olivocerebellar System in Motor Control. Exp. Brain Res. suppl:l7, New York, Springer, pp. 323-327
Simpson, J.l., C.S. Leonard and R.E. Soodak (1988) The accessory optic system of rabbit. II. Spatial organization of direction selectivity. J. Neurophysiol. 60:2055-2072
Simpson, J.l., J. Vander Steen, J. Tan, W. Graf, and C.S. Leonard (1989) Representations of ocular rotations in the cerebellar flocculus of the rabbit. In J.H.J. Allum and M. Hulliger (eds): Progress in Brain Research. Vol. 80. Amsterdam: Elsevier, pp. 213-223
Simpson, J.l., J. Vander Steen, and J. Tan (1992) Eye movements and the zonal structure of the rabbit flocculus. In T. LUnas and C. Sotelo (eds.): The Cerebellum. Springer Verlag, pp. 255-266
Soodak, R.E. and J.I. Simpson (1988) The accessory system of rabbit. I. Basic visual response properties. J. Neurophysiol. 60:2037-2054
Spa<;ek, J. , L. Parizek, and A.R. Lieberman (1973) Golgi cells, granule cells and synaptic glomeruli in the molecular layer of the rabbit cerebellar cortex. J. Neurocyte!. 2:407-428
Stanton. G.B. (1980) Afferents to oculomotor nuclei from area "y" in Macaca mulatta: an anterograde degeneration study. J. Comp. Neurol. 192:377-385
Stroud, B.B. (1895) The mammalian cerebellum. J. Comp. Neurol. 5:71-!18
Takeda, T., and K. Maekawa (1976) The origin of the pretecto-olivary tract. Brain Res. 117:319-325
Takeda, T., and K. Maekawa (1980) Bilateral visual inputs to the dorsal cap of the inferior olive: differential localization and inhibitory interactions. Exp. Brain Res. 39:461-471
Takeda, T, and K. Maekawa (1984a} Collateralized projection of visual climbing fibers to the flocculus and nodulus of the rabbit.
179
Neurosc!. Res. 2:125-132 Takeda. T. and K. Maekawa (1984b) Dorsal cap neurons projecting to the
flocculus, nodulus and uvula with axonal branches. Neurosc!. Lett. Suppl. 17: S73
Takeda. T .. and K. Maekawa (1989a) Olivary branching projections to the flocculus, nodulus and uvula in the rabbit. !. An electrophysiological study. Exp. Brain Res. 74:47-62
Takeda. T .. and K. Maekawa (1989b) Olivary branching projections to the flocculus, nodulus and uvula in the rabbit. II. Retrograde double labeling study Witb fluorescent dyes. Exp. Brain Res. 76:323-332
Tan. J .. L.M. Eisenman. N.M. Gerrits. R. Hawkes. and J. Voogd (1989) Compartmentalization within the rabbit flocculus. A mab Qll3 immunohistochemical, acetylcholinesterase and anatomical tracing study. Eur. J. Neurosc!. Suppl. 2:33
Tan, H., and N.M. Gerrits (1992) Laterality in the vestibulo-cerebellar mossy fiber projection to flocculus and caudal vermis in the rabbit: a retrograde fluorescent double-labeling study. Neuroscience, 47:909-919
Thunnissen, I.E. (1990) Vestibulocerebellar and vestibulo-oculomotor relations in the rabbit. Thesis, Rotterdam.
Thunnissen, I.E., A.H. Epema, and N.M. Gerrits (1989) Secondary vestibulocerebellar mossy fiber projection to the caudal vermis in tbe rabbit. J. Comp. Neurol. 290:262-277
Trojanowski, J.Q., J.O. Gonatas, and N.K. Gonatas (1981) Conjugates of horseradish peroxidase (HRP) with cholera toxin and wheat germ agglutinin are superior to free HRP as orthogradely transported markers. Brain Res. 223:381-385
Trott, J.R., and D.M. Armstrong (1987a) The cerebellar corticonuclear projection from lobule Vb/c of the cat anterior lobe: a combined electrophysiological and autoradiographic study. I. Projection from the intermediate region. Exp. Brain Res. 66:318-338
Trott, J.R., and D.M. Armstrong (1987b) The cerebellar corticonuclear projection from lobule Vb/c of the cat anterior lobe: a combined electrophysiological and autoradiographic study. II. Projection from the vermis. Exp. Brain Res. 68:339-358
Van der Steen, J., and H. Collewijn (1984) Ocular stability in the horizontal, frontal and sagittal planes in the rabbit. Exp. Brain Res. 56:263-274.
Van der Steen, J., J.l. Simpson, and J. Tan (1989) Eye movements evoked from histochemically distinguished compartments of the rabbit flocculus. Soc. Neurosci. Abstr.l5:613. (abstract).
180
Van der Steen. J .. J.I. Simpson, and J. Tan (Hi91) Representation of three-dimensional eye movements in the cerebellar flocculus of the rabbit. In R. Schmid and D. Zambarblert (eds): Oculomotor Control and Cognitive Processes. Amsterdam: ElseVier. pp 63-77
VanRossum, J (1969) Corticonuclear and corticovestibular projections of the cerebellum. Thesis, Leiden. Assen: Van Gorcum
Voogd. J. (1964) The Cerebellum of the Cat. Thesis. Leiden
Voogd. J. (1969) The Importance of fibre connections in the comparative anatomy of the mammalian cerebellum. In R. Llinas{ed): Neurobiology of Cerebellar Evolution and Development. Chicago: A.M.A., pp.493-514
Voogd. J. (1975) Balk's subdivision of the mammalian cerebellum. Growth centres and functional zones. Acta Morpho!. Neerl. Scand. 13:35-54
Voogd, J. (1982) The olivocerebellar projection in the cat. In S.L. Palay and V. Chan-Palay (eds): The Cerebellum: New Vistas. NewYork: Springer Verlag, pp 135-161
Voogd, J. (1989) Parasagittal zones and compartments of the anterior vermis of the cat cerebellum. In P. Strata (ed): The Olivocerebellar System in Motor Control. Exp. Brain Res. Suppl.17. New York: Springer Verlag, pp 3-19
Voogd. J .. and F. Bigare. 1980. Topographical of olivary and corticonuclear fibers in the cerebellum. A reView. In J .Courville, C.C. de Montigny. and Y. Lamarre (eds): The Inferior Olivary Nucleus: Anatomy and Physiology. New York: Raven Press,pp. 207-234.
Voogd. J. and H.K.P. Feirabend (1981) Classic methods in neuroanatomy. In R. Lahue (ed): Methods in Neurobiology, Vol. 2. New York: Plenum, pp 207-234
Voogd, J .. N.M. Gerrits, and E. Maran! (1985) The cerebellum of the rat. In G. Paxinos (ed.): The Rat Nervous System. Vol. 2: Hindbrain and Spinal cord. London, Academic Press, pp. 251-291
Voogd, J .. N.M. Gerrits. and D.T. Hess (1986) Parasagittal zonation of the cerebellum in macaques: an analysis based on acetylcholinesterase histochemistry. In M. Glickstein, C. Yeo.and J. Stein (eds): Cerebellum and Neuronal Plasticity. New York: Plenum. pp. 15-39
Voogd. J .. D.T. Hess, and E. Maran! (1987) The parasagittal zonation of the cerebellar cortex in cat and monkey. In J.S.King (ed): New Concepts in Cerebellar Neurobiology. Neurology and Neurobiology Vol. 22. New York: Alan R. Liss Inc., pp. 183-220
Walberg. F .. N. Kotchabhakdi and G.H. Hoddevik (1979) The olivocerebellar projections to the flocculus and paraflocculus inthe cat, compared to those in the rabbit. A study using horseradish peroxidase as tracer. Brain Res. 161:389-398
Walberg. F., T. Nordby. and E. Dietrichs (1987) The olivonodular projection: a re-examination based on folial cerebellar implants .. Neurosci. Lett. 81:82-87
181
Whitworth. R.H. Jr., D.E. Haines. and G.W. Patrich (1983) The inferior olive of a Prosimian primate. Galago senegalensis. II. Olivocerebellar projections to the vestibulocerebellum. J. Comp. Neurol. 219:228-240
Yamamoto. F., Y. Sato, and T. Kawasaki (1986) The neuronal pathway from the flocculus to the oculomotor nucleus: an electrophysiological study of group y nucleus in cats. Brain Res. 3 71:350-354
Yamamoto. M. (1978) Localization of rabbit's flocculus Purkinje cells projecting to the lateral cerebellar nucleus and the nucleus prepositus hypoglossi investigated by means of the horseradish peroxidase retrograde axonal transport. Neurosci. Lett. 7:197-202
Yamamoto, M. (1979a) Topographical representation in rabbit cerebellar flocculus for various afferent inputs from the brainstem investigated by means of retrograde axonal transport of horseradish preoxidase. Neurosci. Lett. 12:29-34
Yamamoto, M (1979b) Vestibula-ocular reflex pathways of rabbits and their representation in the cerebellar flocculus. In Granit, R, Pompeiano, 0 (eds.) : Reflex Control of Posture and Movement. Progr. Brain Res. 50: 451-457 Amsterdam: ElseVier/North-Holland.
Yamamoto, M., and I. Shimoyama (1977) Differential localizationof rabbit's flocculus Purkinje cells projecting to the medial and superior vestibular nuclei, investigated by means of the HRP retrograde axonal transport. Neurosci. Lett. 5: 279-283
Yamamoto, M., I. Shimoyama and S.M. Highstein (l 978) Vestibular nucleus neurons relaying excitation from the anterior canal to the oculomotor nucleus. Brain Res. 148: 31-42
Yamamura. H.l., and S.H. Snyder (1974) Muscarinic cholinergic binding in rat brain. Proc. Nat!. Acad. Sci. 71:1725-1729
182
8. Summery
This thesis describes the subdivision of the white matter of the
rabbit flocculus and the nodulus on the basis of acetylcholinesterase
(AChE) histochemistry (chapter 1). the compartmental organization of
afferent pathways from the inferior olive to the flocculus and nodulus
(chapter 2), the efferent pathways from the zonally organized Purkinje
cells of the flocculus to the vestibular and cerebellar nuclei (chapter 3).
and the characteristics of eye movements evoked by microstimulation of
the rabbit flocculus {see chapter 4).
The flocculus of the cerebellum is involved in eye movement
control. Several anatomical and electrophysiological studies have already
indicated that the flocculus is subdivided into zones, each of which
controls specific eye muscle activity. However, the zones delineated in
the various studies differ in number and extent. Some of these differences
are undoubtedly due to indiVidual variations in size and number of the
floccular folia. If an internal frame of reference was available to compare
the histological and physiological data from Individual experiments, then
the problems caused by the IndiVIdual variations In the morphology of the
flocculus would be solved.
Our study has demonstrated that AChE reveals the presence of
raphes (sheets of thin myelinated fibers) that subdivide the cerebellar
white matter into compartments containing coarser fibers. The
compartments in the white matter of the flocculus were studied in
transverse, horizontal and sagittal sections processed for AChE. This survey included the morphology of the nodulus, the ventral paraflocculus.
the floccular peduncle and the associated vestibular and cerebellar nuclei (chapter 1).
The granular layer stains most densely with AChE, while the
Purkinje cells are AChE-negative. The molecular layer consists of two
regions. The inner region, located close to the Purkinje cell layer stains
more densely for AChE than the outer two-thirds of the molecular layer
which hardly contains any AChE. This subdivision of the molecular layer is
characteristic for the flocculus and the ventral paraflocculus and it
183
suddenly disappears at the border of the ventral and dorsal paraflocculus.
The white matter of the flocculus is subdiVided by 4 raphes into 5
compartments, called C2, FC 1-FC4 from caudolateral to rostromedial.
Medially the white matter of the flocculus is delineated from the
brachium pontls by a densely staining raphe. which at more caudal levels
becomes less clear. The compartments FC1-FC4 blend together caudally
and travel through group y as the floccular peduncle. Group y contains
intensely AChE-stained neurons. It consists of a ventral portion of closely
packed cells, forming a cap over the restiform body and a dorsal more
loosely arranged portion With larger cells. Medially group y merges with
the caudal pole of the superior vestibular nucleus. The ventromedial
extension of the lateral cerebellar nucleus, consisting of small- and
medium-sized cells (Lpc) is located dorsal to the floccular peduncle and
group y.
Despite variations in number and extent of the floccular folia. the
location of the raphes in the floccular white matter follows a strict
pattern. This compartmentation can serve as the intrinsic framework that
is required to accurately match data from anatomical tracing {see
chapters 2 and 3) and electro physiological experiments (see chapter 4)
instead of using the folia and interfolial fissures as landmarks.
The flocculus receives climbing fiber input from the dorsal cap.
ventrolateral outgrowth, and the rostral pole of the medial accessory
olive. Visual signals are relayed via the dorsal cap and adjoining
ventrolateral outgrowth. In chapter 2 WGA-HRP anterograde tracing
combined with AChE histochemistry was used to find out whether the
climbing fiber projection to the flocculus and the nodulus is zonally
organiZed according to the intrinsic reference frame as described in
chapter l. Injections were placed in subsets of olivary neurons, which
were identified by their electrophysiological characteristics.
The afferent climbing fiber bundle to the flocculus separates into
five smaller bundles in the white matter, each of which becomes located
Within a compartment delineated by AChE-positive raphes. The caudal
part of the dorsal cap sends its climbing fibers through compartments
FC2 and FC4 to terminate on Purkinje cells located in longitudinal strips
184
of molecular layer {floccular zones FZn and FZrv respectively}. Afferent
fibers from the rostral part of the dorsal cap and ventrolateral outgrowth
course through compartments FC 1 and FC3 and terminate in FZ1 and
FZnr. Compartment C2 contains climbing fibers from the rostral pole of
the medial accessory olive (rMAO) which terminate in floccular zone C2.
The boundaries between the floccular zones are sharp and are in line with
the raphes. A similar pattern is observed in the nodulus, the white matter
of which is subdiVided into 4 compartments, XC1 -XC4. Injections in the
caudal dorsal cap and subnucleus .13 label fibers in compartments XC 1 and
XC4, which terminate as climbing fibers in the XZ1 and XZrv zones.
Climbing fibers from the rostral dorsal cap and ventrolateral outgrowth
terminate in zone XZn on the ventral and dorsal surface of lobule X and a
central strip in XZrv (called XZrv*) via compartments XC2 and XC4.
Injections involving the rostral MAO label XC3 and XZm on the dorsal side
of lobule X and the ventral side of lobule IX. These results show the close
correspondence between the white matter compartments (FC and XC)
and the Purkinje cell zones (FZ and XZ).
In chapter 3 the efferent projection of the floccular zones to the
vestibular nuclei was studied with HRP retrograde tracing, WGA-HRP
anterograde tracing and AChE histochemistry. As in chapter 2 the
injection site was identified by the physiological properties of the
different Purkinje cell zones. FZr and FZm project Via compartments FC 1
and FC3 to the dorsolateral part of the superior vestibular nucleus. Fibers
from these compartments also appear to terminate on cells of dorsal
group y. FZn and FZrv project to the medial vestibular nucleus via FCz
and FC4. No injections were made in zone C2. The results are compared
to observations in the preVious chapter.
Chapters l-3 show that specific subsets of olivary neurons are
connected with specific Purkinje cell zones, which in turn project to
specific parts of the vestibular nuclei. Furthermore, these zones correlate
well with the AChE white matter compartmentation. This zonally
organized pathway may serve as the basic neuronal circuitry for
185
differential control of eye muscle activity. Irideed. local electrical
stimulation in different parts of the flocculus does elicit specific types of
eye movements. The published results of earlier microstimulation studies
are hampered by the variations in the morphology of the flocculus.
In chapter 4, AChE histochemistry has also been used to verify the
position of stimulation sites marked by small lesions. The three
dimensional, binocular eye movements evoked by electrical micro
stimulation of the cerebellar flocculus of alert. pigmented rabbits were
recorded using the magnetic induction coil technique. Search coils were
implanted on both eyes. A voltage was induced in the coils by placing
them in a alternating magnetic field and changes in the induced voltage
indicated the displacement of the eyes. Eye movements were registered
with reference to an orthogonal coordinate system consisting of a vertical
axis and two horizontal axes at 45° and 135° azimuth. The azimuth
coordinate was taken to increase to both sides from the 0° reference in
the direction of the nose. Eye movements were most readily evoked by
stimulation (0.2 msec pulses at 200 Hz for 1 sec, intensity 20 ~or less)
at loci in the deep granular layer and the white matter. They consisted of
slow (5-20° /sec) movements. The responses occurred either in both
eyes, With the eye ipsilateral to the stimulated flocculus usually having the
largest amplitude, or were restricted to the ipsilateral eye. The evoked
responses were classified according to the combination of the largest
measured component of rotation and its sense of rotation (clockwise, CW,
or counterclockwise, CCW). Seventy-eight percent of the evoked eye
movements could be placed in one of two classes. For one of these
classes, the largest response component was an abduction of the
ipsilateral eye about its vertical axis (19%), whereas for the other class
the largest response component was a CCW rotation of the ipsilateral
(left) eye about its !35° axis (59%).
The two main classes of three-dimensional eye movements are
differentially associated with anatomical compartments revealed by
acetylcholinesterase histochemistry. Eye movements were elicited with
stimulation of the three largest compartments in the floccular white
matter (FC 1. FCz, and FCs). The middle of these three compartments
186
(FCz) is associated With the vertical axis class of movements, whereas the
two adjacent compartments {FC1 and FC3) are associated with the 135°
class of eye movements. The relation of the two other, smaller
compartments (Cz and FC4) was not determined.
The orientations of the rotation axes of the two main classes of eye
movements correspond to those of the preferred axes of the Visual
climbing fiber input to the flocculus, which suggests that both are
organized in a similar coordinate system. The spatial orientation of the
axes of this coordinate system bears a close resemblance to that of the
axes of the semicircular canals and to that of the rotation axes of
particular pairs of eye muscles.
The association of distinct classes of eye movements and climbing
fiber afferents with the anatomically distinguishable compartments
suggests that the rabbit flocculus consists of anatomically definable
modules that constitute the structural correlate of a coordinate system
used by the brain to control compensatory eye movements.
187
188
9. Somenuotting
Dit proefschrift beschrijft een onderzoek naar de indeling van de
witte stof en de schors van de flocculus en de nodulus in het cerebellum
van het konijn op grand van het voorkomen van het enzym
acetylcholinesterase (AChE) (hoofdstuk 1), de organisatie van afferente
verbindtngen van de oliva infertor naar de flocculus en nodulus (hoofdstuk
2), de efferente verbindtngen van de flocculus naar de vestibulaire en
cerebellaire kernen (hoofdstuk 3). en de karakteristieken van
oogbewegingen die door microstimulatie van de flocculus kunnen worden
opgewekt (hoofdstuk 4).
De flocculus van het cerebellum speelt een belangrijke rol bij de
centrale van oogbewegtngen. Diverse anatomische en elektrofysiologische
studies hebben reeds aangetoond. dat de flocculus in anatomisch
afgrensbare zones is verdeeld. Er bestaan echter grate verschillen in
aantal en grootte van deze zones tussen de verschillende studies.
Waarschijnlijk worden deze verschillen in de hand gewerkt door
individuele variaties in het aantal folia en de grootte en vorm van de folia
van de flocculus. Met behulp van een intrinsiek referentie kader zouden
de anatomische en fysiologische data van verschillende experimenten
kunnen worden vergeleken en zouden de problemen, die ontstaan door
de ind!viduele varlatie in de morfologle van de flocculus zijn opgelost.
Hersenmateriaal, dat gekleurd wordt met een enzym
histochemische methode voor acetylcholinesterase (AChE). laat een
bijzonder patroon zien, waarbij dunne raphes, die uit dunne AChE
positieve vezels bestaan, de Witte stof van het cerebellum verdelen in
compartlmenten, waarin dikkere vezels, die niet met AChE kleuren, zijn
gelegen. In hoofdstuk 1 concentreerde het onderzoek zich op de vraag of
de witte stofvan de flocculus in het konijn met behulp van deze techniek
in compartimenten verdeeld zou kunnen worden. Hiervoor werden
hersenen van konijnen in transversale, hortzontale en sagittale coupes
gesneden en gekleurd voor AChE. Het onderzoek richtte zich tevens op
de morfologte van de ventrale paraflocculus, de pedunculus van de
flocculus en daaraan gerelateerde vestibulaire en cerebellaire kernen.
189
De sterkste aankleuring voor AChE wordt gezien in de granulaire laag,
terwijl Purkinje cellen ongekleurd blijven. De moleculaire laag lijkt uit
twee delen te bestaan, waarvan het binnenste eenderde deel. dat tegen de
Purkinje cellaag is gelegen, relatief hard aankleurt in vergelijking met het
buitenste tweederde dee!, dat haast geen AChE bevat. Dit patroon wordt
gevonden in de flocculus en de ventrale paraflocculus en verdwijnt
abdrupt op de grens van de ventrale met de dorsale paraflocculus. De
Witte stof van de flocculus wordt van het brachium pontis gescheiden
door een sterk AChE-positieve raphe, die caudaal minder duidelijk wordt.
De witte stof zelf wordt door 4 raphes verdeeld in 5 compartimenten.
Deze compartimenten worden van caudolateraal naar rostromediaal
benoemd als C2. FC1 -FC4. Caudaal versmelten de compartimenten
FC1-FC4 en vormen zij de pedunculus van de flocculus. Deze verloopt door
groep y van de vestibulaire kern en in de rtchting van de caudate pool van
de nucleus vestibularis superior. De celgroep y bestaat uit een ventrale,
compacte laag van cellen, die een kapje vormen, over het corpus
restiforme, en een dorsale subnucleus met grotere, spoelvormige cellen
tussen de vezels van de pedunculus flocculi. Aile neuronen van de groep y
kleuren sterk voor AChE. Groep y grenst aan de dorsale zijde aan de
ventromediale uitloper van de laterale cerebellaire kern, die kleine en
middelgrote cellen bevat (Lpc). Mediaal versmelt groep y met de superior
vestibulaire kern.
Ondanks individuele verschillen in aantal en grootte van de folia van
de flocculus behouden de raphes in de witte stof van de flocculus hun
positie. Deze onderverdeling in compartimenten wordt in dit proefschrift als intrinsiek referentiekader gebruikt voor de vergelijking van
topografische gegevens. verkregen via anatomische tracing (zie
hoofdstukken 2 en 3) en elektrofysiologisch onderzoek (zie hoofdstuk 4).
Deze methode is veel nauwkeuriger dan de plaatsbepaling ten opzichte
van folia en interfoliale fissuren wals tot dusver gebruikelijk was.
De flocculus ontvangt klimvezels afkomstig van de volgende
subnuclei van de onderste olijf: de dorsal cap, de ventrolateral outgrowth
en de rostrale pool van de mediale bijolijf. De dorsal cap en de
ventrolateral outgrowth ontvangen optokinetische informatie uit de kern
190
van de tractus opticus en de kernen van de accesSoire tractus optlcus in
het mesencephalon. In hoofdstuk 2 werd anterograde tracing met
WGA-HRP. gecombineerd met AChE histochemie. gebruikt om na te gaan
of de klimvezelprojectie naar de flocculus en de nodulus een zonale
indeling kent. die zich houdt aan het in hoofdstuk l beschreven
intrinsieke referentie kader. Bij het maken van de injecties met
WGA-HRP werden de subnuclei van de oliva inferior geidentificeerd aan
de hand van hun fysiologische karaktertstleken.
De hoofdbundel van klimvezels naar de flocculus splitst zich in 5
kleinere bundels. Elk van deze bundels bezet een compartiment, dat
begrensd is door AChE-positleve raphes. Het caudale deel van de dorsal
cap stuurt klimvezels Via compartimenten FC2 en FC4. terwijl afferente
vezels vanaf het rostrale deel van de dorsal cap en de ventrolateral
outgrowth gebruik maken van compartimenten FC1 en FC3.
Compartiment C2 bevat klimvezels afkomstig van de rostrale mediale
bijolijf (rMAO). De vezels binnen elk compart!ment eindigen in een
nauwkeurig af te bakenen longitudinale strook van de moleculaire laag
(flocculus zones C2, FZJ-FZJV). De begrenzing van deze stroken is scherp
en ligt in het verlengde van de raphes. Eenzelfde patroon is aanwezig in
de nodulus met 4 cornpartimenten in de witte stof, cornpartimenten
XC1-XC4. lnject!es in het caudale dee! van de dorsal cap en nucleus J3
laten gelabelde vezels zien in compartimenten XC 1 en XC4, welke als
klimvezels eindigen in zones XZI en XZIV van de nodulus. Klimvezels uit
het rostrale deel van de dorsal cap en de ventrolateral outgrowth eindigen
via compartimenten XC2 en XC4 in zone XZn op het ventrale en dorsale
gedeelte van lobulus X en in een centrale strip binnen zone XZIV·
Injecties, die de rMAO meenemen, sturen hun vezels Via cornpartiment
XCm naar wne XZm op de dorsale zijde van lobulus X en de ventrale zijde
van lobulus IX. Deze bevindingen laten zien dat elk van de witte stof
compartimenten (FC en XC} aan een bepaalde Purkinje eel zone (FZ en
XZ) is gerelateerd.
In hoofdstuk 3 werd de efferente projectie van de zones in de
flocculus naar de vestibulaire kernen bestudeerd met behulp van HRP
191
retrograde tracing, WGA-HRP anterograde tracing, en AChE histochemie.
Evenals in hoofdstuk 2 werden de injectieplaatsen bepaald aan de hand
van de fysiologische karakteristieken van de verschillende Purkinje cell
zones. De zones FZI en FZni sturen axonen Via compartiment FC1 en FC3
naar het dorsolaterale gedeelte van de nucleus vestibularis superior.
Vezels behorende bij deze compartimenten lijken oak te eindigen in
groep y. Op vergelijkbare wijze projecteren FZu en FZrv naar de mediale
vestibulaire kern via FC2 en FC4. Geen van de injecties was gelokaliseerd
binnen C2. De resultaten werden vergeleken met die van het vorige
hoofdstuk.
In hoofdstukken 1-3 wordt beschreven hoe subgroepen binnen de
ohva inferior in contact komen met specifieke Purkinje eel zones, die op
hun beurt projecteren naar subgroepen binnen de vestibulaire kernen.
Deze zones zijn morfologisch geassocieerd met specifieke
compartimenten in de witte stof. Deze zonale verbindingen zouden het
neuronale substraat kunnen zijn van gedifferentieerde controle over
oogspier activiteit door de flocculus. La kale elektrische stimulatie in de
flocculus veroorzaakt verschillende typen oogbewegingen. Alweer blijkt,
dat een consistente interpretatie van de resultaten van dergelijke
experimenten verstoord wordt door de variabele morfologie van de
flocculus. Zodoende werd AChE histochemie ook in hoofdstuk 4 ingezet
om de relatieve plaatsing van de stimuluslocaties, die gemerkt werden
door kleine lesies, te verifH!ren.
De drie-dimensionale, binoculaire oogbewegingen. die door
electrische microstimulatie van de flocculus van het cerebellum in
wakkere gepigmenteerde konijnen konden worden opgewekt, werden
geregistreerd met behulp van de magnetische inductie methode. Op
beide ogen werden inductiespoeltjes geimplanteerd op de sclerae.
Geplaatst in een homogeen magnetisch veld zal er bij een oogbeweging
een spanning worden gelnduceerd binnen de spoeltjes. Veranderingen in
deze spanning zijn direct gerelateerd aan de grootte en de richting van
de oogbeweglng.
Voor de beschrijving van de oogbewegingen werd een orthogonaal
192
referentiekader bestaande uit een verticale as en twee horizontale assen
gebruikt. De hor:tzontale assen maken een hoek van 45° en 135° met de
rostraal gerichte 0° as.
Oogbewegingen konden het gemakkelijkst worden opgewekt door
te stimuleren {0,2 msec pulsen bij 200 Hz gedurende 1 sec, :::;; 20 ~) in
de diepe granulaire laag en de Witte stof. De oogbewegingen waren vrij
langzaam (5-20'/sec). Be!de ogen reageerden op de stimulus. waarblj het
oog, dat ipsilateraal was gelokaliseerd ten opzichte van de gestimuleerde
flocculus. doorgaans de grootste amplitude had. Soms waren de
oogbeweg!ngen ultslu!tend !ps!lateraal gelokal!seerd. De oogbewegingen
werden geclassificeerd aan de hand van de grootste oogbewegtngs
component en de rtchting van de rotatie {met de klok mee, CW, of tegen
de klok in, CCW). Achtenzeventig procent van de geregistreerde
oogbewegingen kon worden geplaatst in een van twee groepen. In een van
de groepen bestond de grootste respons uit een abductie van het
ipsilaterale oog random de verticale as (19%), terw:ijl voor de andere
groep (59%) de grootste component een CCW rotatie was van het
ips!laterale (linker) oog random de !ps!laterale 135' as.
Deze twee voornaamste groepen van drie-dimensionale
oogbewegingen zijn nauw geassocieerd met de anatomische
compartimenten, die met behulp van AChE histochemie kunnen worden
begrensd. Voor de drte grootste van de v!Jf compart!menten (FC1. FCz en
FC3) z!jn stlmulatle gegevens besch!kbaar. De m!ddelste van deze 3 (FC2)
is gerelateerd aan de eerste groep. met rotaties rand de verticale as, de
twee flankerende compartimenten (FC 1 en FCs) zijn betrokken bij
rotaties rand de 135° as uit de tweede groep. De functionele betekenis
van de twee andere, kleinere compartimenten (C2 en FC4) kon niet
worden vastgesteld.
De orientatie van de rotatie-assen voor oogbewegingen, opgewekt
door stimulatie van een bepaald compartiment. komt overeen met de
rtchting van de rotatie-assen voor bewegingen van de omgeving. die een
optimale respons uitlokken in de klimvezels die eindigen op Purkinje
cellen van de corresponderende zone. Dit suggereert, dat de flocculus uit
193
modules is opgebouwd die voor hun klimvezel input en hun output
hetzelfde coordinaten systeem hanteren. De positie van deze assen komt
geed overeen met de positie van de assen van de halfcirkelvormige
kanalen en met die van bepaalde oogspier paren. Elke module, met zijn
eigen set van aanvoerende en afvoerende verbindingen, representeert
oogbewegingen rand een bepaalde as.
Het verband, dat werd aangetoond tussen bepaalde klassen
oogbewegtngen, klimvezel afferenten en Purkinje eel efferenten met
anatomische compartimenten, maakt het waarschijnlijk dat de flocculus
uit modules bestaat, die het structurele correlaat vormen van een
coordinaten systeem, dat door de hersenen wordt gebruikt bij de
controle van compensatoire oogbewegingen.
194
I 0. Dani<woord
Bij deze wil ik degenen bedanken, die de basis vormden van dit proefschrift.
Mijn vader, die tegen een ziekte opliep, die mijn interesse in hersenonderzoek opwekte. Zijn strijd tegen ziekte en zijn belangstelling voor onderwtjs zullen in mij voortleven. Mijn moeder met haar nimmer aflatende belangstelling en aanmoedigingen als er weer eens een dip naderde, een echte moeder.
Mijn promotoren, Beste Jan, na een sputterende start en een langdurige inhaalmaneuvre is het er tach nag van gekomen. Je enorme anatomische kennis en niet te evenaren tekenkunsten zullen mlj voor altijd bijblijven. Dear Jerry, this research project gained momentum after you put your weight behind it. Without your physiological knowledge and your Willingness to invest time to study the anatomical side, this project would have been less "functional". Your love for big and small ants stretched my patience, but really polished up the goods. I am glad we shared other interests, such as the visual, olfactory, and gustatory input to the brain. I hope we will be able to continue sharing these life events in the future. Beste Hans. hoofdstuk 4 zou niet bestaan zonder jouw technische kwaliteiten, je nuchtere aanpak en de vele slaapverwekkende uren die we in het danker samen met Jerry hebben doorgebracht. Beste Henk. dank voor het plaats nemen in de leescommissie en voor je suggesties om tegenstrijdigheden uit de weg te ruimen.
De analisten. Eddie "De Walm" Dalm, zonder jouw operatieve en fotografische kwaliteiten en bereidheid om buiten kantooruren door te werken zou dit proefschrift nog jaren op zich hebben laten wachten. Meestal werden deze kwaliteiten begeletd door rookwolken. Zodoende zou ik zeggen. tot ziens bij de KNO. Hans van der Burg, je geduld en technisch vernuft heeft ervoor gezorgd, dat peperdure onwerkzame fabrieksapparatuur vervangen kon worden door handzame eigenfabricaten. die wel voldoen. Zander jouw "puff geval zouden hoofdstukken 2 en 3 niet geschreven zijn. Rob Boer. jouw speurtocht naar de juiste verhoudingen en ingredienten stand aan de basis om de AChE histochemie en de HRP-gel techniek van de grand te krijgen. Je plakwerk was perfect. Hennie K!Jnk, Peter Elffer!ch, Erika Goedknegt, ju111e hebben samen met Rob tienduizenden coupes gesneden, geTMBd. geDABd, geAChEd. geHaggqvist, gedroogd, opgeplakt, ontwaterd. tegengekleurd met cresylviolet. neutraal rood, gexylold, gePermount, afgedekt enz. enz ......... . De ware ruggegraat van dit proefschrift ligt in deze handelingen. Paula van Alphen, de creatieve fotografe wtens werk altijd sublieme kwaliteit vertoonde Corlinda Bongers, een te WOW afscheidscadeau.
195
Edith Klink, de elegante en charmante spil vim de afdeling. Jouw typewerk en flexlbele geest heeft me gelukkig vee! tijd opgeleverd (die ik anders onder andere kwijt was geweest aan het ontcijferen van Jan's handschrift).
Gert Holstege, samen met Wijlen Prof. H. Kuijpers, heeft hij mij toen ik studentassistent neuro-anatomie bij hem was, in de eerste plaats enthousiast gemaakt over het fenomeen hersenen. Zijn manier van lesgeven maakt hem uniek. wat deels in mijn stijl van onderwijs geven nog altijd doorklinkt.
Joan Holstege, als ik een beetje in de put zat was je er. Dank je voor je gezelschap. zeker voor de uren buiten de gebruikelijke kantoortijden. Dr L. Eisenman, dear Len, I would like to thank you for your hospitality during my stay at your department in Philadelphia. It was the feeding time of my life. I Will be back.
Verder wil ik alle vrienden en kennissen bedanken voor hun aandacht en steun, in het bijzonder Judith (paraniml], Thea (paraniml), Jacqueline, Rob, Harold, Astrid, Steven Jobs, Enzo Ferrari. Antonio Carlos Jobim en Star Trek.
Niet bedankt worden de makers van Trtstan Pinball, Space Quest, Wolf3D, Sub, Spectre, Crystal Quest, Paratrooper, PacMan, Apache Strtke Mission, F-16 Falcon, Chessmaster, The Fool's Errant, MS-DOS enz ....
196
11. Curriculum uitae
De auteur van dit proefschrift werd geboren op 20 september 1962 te Jogjakarta in Indonesit~. Het eindexamen gymnasium J3 werd afgelegd
in 1980 aan het Erasmiaans gymnasium te Rotterdam. Vervolgens studeerde hiJ geneeskunde aan de Erasmus Universiteit Rotterdam,
alwaar in !985 het doctoraal examen werd behaald. In 1983 koos hij
neuro-anatomie als keuzeprakticum en bleef tot 1986 als student
assistent/wetenschappelijk onderzoeker aan de afdeling Anatomie
verbonden. Vanaf 1986 tot 199! was h!J in dienst als AIO (assistent in
opleiding) bij deze afdeling, waar het onderzoek werd verricht dat tot dit
proefschrift heeft geleid. In de periode 1991-1992 doorliep hij alsnog de
senior co-schappen. Dit werd afgerond met het behalen van de artsenbul
in juli 1992. Sinds september 1992 is hij werkzaam als arts-assistent op
de afdeling KNO/Hoofd-hals chirurgie van het Academisch Medisch
Centrum te Amsterdam (sinds oktober 1993 in opleiding).
197
12. Ust of publications
- Mesencephalic projections to the facial nucleus in the cat. An autoradiographical tracing study. G. Holstege, J. Tan, J. J. van Ham. and A. Bos. Brain Research 311 (1984) 7-22
- Afferent projections to the orbicularis oculi motoneuronal cell group. An autoradiographical tracing study in the cat. G. Holstege, J. J. van Ham. and J. Tan. Brain Research 374 (1986) 306-320
- Anatomical observations on the afferent projections to the Tetractor bulbi motoneuronal cell group and other pathways possibly related to the blink reflex in the cat. G. Holstege, J. Tan, J.J. van Ham, and G. A. Graveland. Brain Research 374 (1986) 321-334
-Anatomical eVidence that the pontine lateral tegmental field projects to lamina I of the caudal spinal trigeminal nucleus and spinal cord and to the Edinger-Westphal nucleus in the cat. J. Tan and G. Holstege. Neuroscience Letters 64 { 1986) 317-322
- Supraspinal control of motoneurons innervating the striated muscles of the pelvic floor including urethral and anal sphincters in the cat. G. Holstege and J. Tan. Brain 110 (1987) 1323-1344
- Anatomic observations on afferent projections of orbicularis oculi and retractor bulbi motoneuronal cell groups and other pathways possibly related to the blink reflex in the cat. G. Holstege. J. Tan, and J.J. van Ham. In: Cellular Mechanisms of Conditioning and Behavioral Plasticity (C.D. Woody, D.L. Alkon, and J.L. McGaugh eds). New York, Plenum Publishing Corporation, pp. 273-286 (1988)
Projections from the red nucleus and surrounding areas to the brain stem and spinal cord in the cat. An HRP and autoradiographical tracing study. G. Holstege and J. Tan. Behavioural Brain Research 28 (1988) 33-57
Representations of ocular rotations in the cerebellar flocculus of the rabbit. J.l. Simpson, J. van der Steen, J. Tan. W. Graf. and C.S. Leonard. In: Progress in Brain Research Volume 80 (J.H.J. Allum and M. Hillliger (eds). Amsterdam, ElseVier. pp. 213-223 {1989)
Representation of three-dimensional eye movements in the cerebellar flocculus of the rabbit. J. van der Steen, J.L Simpson, and J. Tan. ln: Oculomotor Control and Cognitive Processes (R. Schmid and D. Zambarbierl eds). Amsterdam, Elsevier, pp 63-77
- Eye movements and the zonal structure of the rabbit flocculus. J.I. Simpson, J. van der Steen, and J. Tan In: Cerebellum Revisited (R. Llinas. C. Sotelo eds). Springer, Heidelberg, pp. 255-266 (1992)
198
- Reference frames of the rabbit flocculus. J.I. Simpson. J. van der Steen. and J. Tan In: Vestibular and Brain Stem Control of Eye, Head, and Body Movements [H. Shlmazu, Y. Shinoda, eds.). Karger, Tokyo. pp.301-313 (1992)
Binding of the ligand 3H-MK 801 to the glutamate NMDA receptor during acute hepatic encephalopathy and during ammonia infusion simulating acute hepatic encephalopathy. R.J. de Knegt, S.W. Schalm, J. Komhuber, P. Riederer, and J. Tan. Metabol. Brain Dis. 8:81-94
chapter 1= White matter compartmentation in the flocculus of the rabbit. An acetylcholinesterase histochemical study. J. Tan, J.l. Simpson, and J. Voogd J. Camp. Neural. submitted
chapter 2= Zonal organization of the climbing fiber projection to the rabbit flocculus. J. Tan, J.I. Simpson, R. Nanhoe, N.M. Gerrits, and J. Voogd J. Camp. Neural. submitted
chapter 3= Zonal organization of the floccule-vestibular nucleus projection in the rabbit. J. Tan, J.I. Simpson, N.M. Gerrits, and J. Voogd J. Camp. Neural. submitted
chapter 4= The functional and anatomical organization of threedimensional eye movements in rabbit cerebellar flocculus. J. van der Steen, J .I. Simpson, and J. Tan. J. Neurophysiol. submitted
199
200